SPOTLIGHT | 信息 |
ScaleFlux News News 信息 ScaleFlux Events Events 展览日历 ScaleFlux Blog Blog 博客
ScaleFlux™ Wins Best-of-Show Award in Partnership with Alibaba Cloud at Flash Memory Summit 2019
ScaleFlux™ 与阿里云协作荣获2019闪存峰会 “最佳表现奖”
Industry-Wide Focus on Computational Storage at Flash Memory Summit 2019 在2019年闪存峰会上, 可计算存储受到了全球行业关注
August 8th, 2019 | 2019年8月8日 |
via BusinessWire BusinessWire 网站

SANTA CLARA, CA – ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, is proud to receive a Flash Memory Summit best-of-show award in collaboration with partner Alibaba Cloud, the data intelligence backbone of Alibaba Group. Alibaba Cloud and ScaleFlux’s win of the Most Innovative Flash Memory Customer Implementation category comes from the achievement of integrating Computational Storage in Alibaba Cloud’s cloud-native database POLARDB.

Cost-effectively deploying analytical database workloads at cloud scale is a major infrastructure challenge given the immense compute and storage resources required. Alibaba Cloud’s cloud-native database POLARDB has been optimized to capitalize on the merits of ScaleFlux Computational Storage Drives (CSD) to carry out intensive data analysis tasks, pushing the processing down to where the data is stored. This method eliminates massive amounts of data movement (returning pre-processed results instead of moving the whole data set), saving power and latency. Workload parallelization across CSDs eliminates the CPU bottleneck, resulting in scalable performance improvements.

“Computational Storage is now proving its ability to transition from a niche technology to becoming a mainstream solution driving the database application workloads of production database systems at hyperscale customer environments,” said Jay Kramer, Chairman of the Awards Program and President of Network Storage Advisors Inc. “We are proud to recognize Alibaba Cloud’s cloud-native database POLARDB with ScaleFlux Computational Storage to carry out intensive analytical data processing as a key infrastructure solution.”

ScaleFlux’s award win is accompanied by excellent traffic at its premier expo booth and a well-attended Wednesday keynote delivered on the main stage of Flash Memory Summit.

“It is my pleasure and privilege to accept this best-of-show award on behalf of the ScaleFlux engineering team,” said Yang Liu, ScaleFlux co-founder and chief architect. “Our work with Alibaba Cloud demonstrates the value that Computational Storage serves in performance-critical database applications.”

“We are glad to be recognized for this collaboration with ScaleFlux,” said Zhu Feng, Senior Expert of Alibaba Cloud Intelligence. “Alibaba Cloud is working closely with our partner to co-develop and fully utilize Computational Storage to achieve cost-effective and scalable real-time data analytics, all the while delivering responsive cloud-native user experiences.”

ScaleFlux continues to refine its key technology with targets on widespread usage. The newly released CSD 2000 Series brings transparent data path compression/decompression, and TCO savings to enterprise, cloudscale, and hyperscale customers worldwide.

ScaleFlux公司-大规模部署可计算存储方面的先驱,非常荣幸与阿里巴巴集团的数据智能支柱—阿里云合作。双方的合作荣获2019 闪存峰会最佳荣誉奖- “最具创新的闪存客户实践 “,表彰阿里云与ScaleFlux率先在阿里云的云原生数据库POLARDB中独创实现了可计算存储方案。

数据库需要强大的计算能力和存储性能,特别是在云级别部署时数据分析的工作负载对基础设施(计算、存储和网络)来说是巨大的挑战。阿里云的云原生数据库POLARDB针对此难点进行了优化,利用ScaleFlux可计算存储设备 (CSD)来执行密集的数据分析任务,将数据处理工作分流至CSD。这种方式消除了大量的数据移动, 仅仅返回预处理结果, 而不需要在CPU、内存、存储及网络之间移动全部数据,节省了CPU消耗并降低了数据延迟时间。多个CSD同时并行工作,消除了CPU计算能力瓶颈,从而大幅度提高了节点性能。

闪存峰会主席兼互联网存储咨询公司总裁Jay Kramer表示,我们正在见证可计算存储从一项小众技术转变为主流厂商客户环境下实际数据库生产环境负载的主流解决方案。 我们很自豪地认可:阿里云的云原生数据库POLARDB结合了ScaleFlux的可计算存储能力,它可以进行密集的分析数据处理,是一个关键的基础设施解决方案。

“我很高兴也很荣幸能代表ScaleFlux工程团队接受这个最佳表现奖,”ScaleFlux的联合创始人兼首席架构师-刘洋说。“我们与阿里云的合作展示了可计算存储在性能关键的数据库应用程序中的价值。”

阿里云智能资深专家刘振军表示“我们正在与合作伙伴密切协作,以实现具有最佳成本效益和可扩展的实时数据分析,同时提供快速响应的云原生用户体验。”

全新CSD2000系列产品发布

ScaleFlux在2019年的闪存峰会上发布了备受期待的可计算存储驱动器2000系列产品,扩充了现有产品线。新平台的亮点主要有:

  • • 领先的OLTP数据库性能和混合读/写工作负载
  • • 透明的数据路径压缩/解压
  • • 与类似的PCIe SSDs相比,每GB成本最低

ScaleFlux市场与运营副总裁Thad Omura说: “我们很高兴能将可计算存储技术引入主流企业。我们最新的CSD 2000系列产品将为新老客户带来巨大的价值,并以卓越的性能和最优的成本巩固可计算存储的声誉。”

聚焦可计算存储

在过去的一年中可计算存储的发展受到了闪存行业的广泛关注。闪存峰会是全球最大的年度行业盛会,今年的峰会上组织者安排了一整天时间分组讨论可计算存储并召开专题研究。 ScaleFlux作为可计算存储领域的行业翘楚,是首家推出可计算存储产品的公司、2018年闪存峰会最具创新初创企业奖获得者、存储行业协会可计算存储标准委员会的创始成员之一。

ScaleFlux Announces Computational Storage Drive 2000 Series
ScaleFlux Announces Computational Storage Drive 2000 Series
Industry-Wide Focus on Computational Storage at Flash Memory Summit 2019 Industry-Wide Focus on Computational Storage at Flash Memory Summit 2019
August 6th, 2019 | 2019年8月6日 |
via BusinessWire BusinessWire 网站

SANTA CLARA, CA – ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, unveiled today its Computational Storage Drive (CSD) 2000 Series family of products at Flash Memory Summit 2019 (FMS), the industry’s premier annual conference. The breakthrough CSD 2000 Series platform follows the successful deployment of its preceding CSS 1000 Series to bring top-tier compute and storage performance to cloudscale and enterprise markets worldwide.

Computational Storage is at the forefront of increasing industry interest at FMS, commanding a dedicated day of focused panels and technical sessions. Over a dozen companies will present on the technology that looks poised to become the key building block for modern data center infrastructure as massive zettabyte-scale data growth collides with demand for real time analytical processing. ScaleFlux is a founding member of the standards committee SNIA Computational Storage TWG and will be on deck to moderate discussion as well as present on its own technology breakthroughs.

Thad Omura, EVP of Marketing and Operations for ScaleFlux, will deliver a keynote presentation titled “Data-Driven Applications Gravitate to Computational Storage” on Wednesday, Aug 7 @ 1:30pm in the main hall of Flash Memory Summit. This anticipated talk will set the stage for the present development of Computational Storage as well as deliver the bulk of detail on the CSD 2000 Series, which promises huge across-the-board application performance benefit from functions such as transparent inline compression/decompression.

ScaleFlux carries much momentum from the noteworthy adoption of its CSS 1000 Series, which has shipped thousands of units worldwide. The product family is uniquely capable of feature upgrades through unobtrusive in-place updates due to its FPGA approach and has continually improved its lifetime value. CSS 1000 Series is built upon state-of-the-art TLC NAND from multiple vendors and is supported by several tier-1 OEMs, including Dell EMC.

NEW CSD 2000 SERIES FAMILY OF PRODUCTS LAUNCHED

ScaleFlux has released its highly anticipated Computational Storage Drive 2000 Series family of products to complement its existing offerings. The new platform is highlighted by:

  • Leading performance for OLTP databases & mixed read/write workloads
  • Transparent inline Compression/Decompression
  • Lowest cost per effective GB, compared to similar user capacity NVMe SSDs

More highlights and implementation details will be unveiled at the ScaleFlux FMS Keynote on August 7th at 1:30pm.

“We are excited to lead Computational Storage to mainstream enterprise adoption,” said Thad Omura. “Two years ago we coined the term, last year we made our industry debut, and now we are at the front of a very large crowd. Our newest CSD 2000 Series products will deliver immense value to new and existing customers alike and cement the reputation of Computational Storage with superior performance and optimal total cost of ownership.”

COME VISIT SCALEFLUX @ FMS 2019

ScaleFlux is a premier sponsor of Flash Memory Summit 2019, held at the Santa Clara Convention Center on August 6-8th. The ScaleFlux executive team and engineering/business staff will be present to discuss the latest market applications and product solutions in Computational Storage highlighted by the following:

  • Jam-packed expo display: Booth #113 during exhibition hours with live demonstrations and exciting giveaways; Dell EMC ESI is a partner hosted at our booth and will display server hardware containing the CSS 1000 Series cards
  • Conference keynote presentation: Thad Omura (EVP of Marketing and Operations for ScaleFlux) will be onstage to deliver: “Data-Driven Applications Gravitate to Computational Storage”, Wednesday, Aug 7 @ 1:30pm in the main hall
  • Benchmark demos: Our applications engineers will be present with multiple demos showcasing application performance benchmarks
  • Panel session: JB Baker (Senior Director of Product Management) will present at Panel Session (SSDS-102-1): “Overcoming the Reductions in NAND Endurance Ratings” on Tuesday, Aug 6 @ 3:40pm as part of the Enterprise SSDs Track
  • Panel session: Tong Zhang (ScaleFlux Co-Founder and Chief Scientist) will present at Technical Session (COMP-301B-1): “Bring Intelligence to Your Database Storage” on Thursday, Aug 8 @ 9:45am as part of the Computational Storage Track
  • Private meeting room: On-site scheduled meetings with customers, partners, press and analysts (email info@scaleflux.com if interested)
Watch the OCP Panel on Computational Storage!
Watch the OCP Panel on Computational Storage!
open compute project global summit 2019 open compute project global summit 2019
March 21st, 2019 | 2019年3月21日 |
via YouTube YouTube 网站

Computational Storage was the talk of the show at OCP Global Summit. Watch the well-attended Q&A panel featuring ScaleFlux's EVP Thad Omura here!

Computational Storage was the talk of the show at OCP Global Summit. Watch the well-attended Q&A panel featuring ScaleFlux's EVP Thad Omura here!

ScaleFlux Computational Storage Elevates Application Performance with Atomic Write and Namespace Streams
ScaleFlux Computational Storage Elevates Application Performance with Atomic Write and Namespace Streams
New Features Debut in 2nd Generation Product Update at Open Compute Project Global Summit 2019 New Features Debut in 2nd Generation Product Update at Open Compute Project Global Summit 2019
March 12th, 2019 | 2019年3月12日 |
via BusinessWire BusinessWire 网站

SAN JOSE, Calif. --(BUSINESS WIRE)-- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, announced today new features in its CSS 1000 product line: Atomic Write and Namespace Streams. ScaleFlux will highlight these product updates at OCP Global Summit San Jose, the premier data infrastructure conference, on March 14-15. These features are available to existing customers (via a downloadable package) and to new customers. Both Atomic Write and Namespace Streams enable performance and endurance benefits for data-intensive workloads.

Atomic write (AW) is the ability to guarantee a single write of blocks up to 1MB in size. For database applications, this assures each Database record update is written in its entirety in a single transaction rather than breaking up the write into multiple transactions. This bypasses the traditionally used Double Write Buffer and increases drive endurance as a result of reducing total write operations. AW preserves existing benefits such as recovering data in the event of a crash and increases transactions per second for write-intensive workloads.

Namespace Streams (NS) greatly improves Flash storage I/O performance, latency and endurance by minimizing unnecessary garbage collection. NS allows customers to create separate Streams of logically-associated data with different life cycles by simply associating each Stream with a Namespace. For example, log updates, files, and transaction records could each be a Stream. CSS then places the data from a given Stream together on physically-associated region of Flash, each with its own set of garbage collection units and its own tunable overprovisioning. This yields lower overall latency and write amplification.

“Large scale data centers are now demanding hardware-based application acceleration faster than their infrastructure refresh cycles,” said Iqbal Bhalwani, VP of Global Sales for ScaleFlux. “Our agile FPGA-based approach allows for tunable performance and deployment of new features to existing production clusters through simple, in-system updates, extending the value of customers’ Computational Storage infrastructure well beyond traditional lifecycles.”

ScaleFlux CSS 1000 Series drives are in volume deployment worldwide among hyperscale, webscale, and enterprise customers across diverse use cases such as fraud detection, mobile payments, AI, gaming, and media delivery. ScaleFlux aims to bring the performance acceleration of Computational Storage to an ever-expanding set of data-driven applications that are increasingly both compute and storage I/O bound.

SCALEFLUX WILL BE AT OCP GLOBAL SUMMIT 2019

In a nod to Computational Storage’s increasing traction in the storage industry as a whole, Open Compute Project will host a panel devoted to the technology at its Global Summit held in San Jose. Thad Omura, EVP Marketing of ScaleFlux, will join representatives from six fellow companies to discuss the present state and future development of Computational Storage. The panel will be held on Friday, March 15th, at 4pm in hall 212 AB.

ScaleFlux will additionally be present in SNIA’s booth #A22. The industry standardization authority has made it a priority this year to feature the new Technical Work Group for Computational Storage, for which ScaleFlux is a founding member. ScaleFlux has a private meeting room for face to face discussion with customers, analysts, and partners, and press. (email info@scaleflux.com if interested in hearing about the latest update in Computational Storage)

SAN JOSE, Calif.--(BUSINESS WIRE)--ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, announced today new features in its CSS 1000 product line: Atomic Write and Namespace Streams. ScaleFlux will highlight these product updates at OCP Global Summit San Jose, the premier data infrastructure conference, on March 14-15. These features are available to existing customers (via a downloadable package) and to new customers. Both Atomic Write and Namespace Streams enable performance and endurance benefits for data-intensive workloads.

Atomic write (AW) is the ability to guarantee a single write of blocks up to 1MB in size. For database applications, this assures each Database record update is written in its entirety in a single transaction rather than breaking up the write into multiple transactions. This bypasses the traditionally used Double Write Buffer and increases drive endurance as a result of reducing total write operations. AW preserves existing benefits such as recovering data in the event of a crash and increases transactions per second for write-intensive workloads.

Namespace Streams (NS) greatly improves Flash storage I/O performance, latency and endurance by minimizing unnecessary garbage collection. NS allows customers to create separate Streams of logically-associated data with different life cycles by simply associating each Stream with a Namespace. For example, log updates, files, and transaction records could each be a Stream. CSS then places the data from a given Stream together on physically-associated region of Flash, each with its own set of garbage collection units and its own tunable overprovisioning. This yields lower overall latency and write amplification.

“Large scale data centers are now demanding hardware-based application acceleration faster than their infrastructure refresh cycles,” said Iqbal Bhalwani, VP of Global Sales for ScaleFlux. “Our agile FPGA-based approach allows for tunable performance and deployment of new features to existing production clusters through simple, in-system updates, extending the value of customers’ Computational Storage infrastructure well beyond traditional lifecycles.”

ScaleFlux CSS 1000 Series drives are in volume deployment worldwide among hyperscale, webscale, and enterprise customers across diverse use cases such as fraud detection, mobile payments, AI, gaming, and media delivery. ScaleFlux aims to bring the performance acceleration of Computational Storage to an ever-expanding set of data-driven applications that are increasingly both compute and storage I/O bound.

SCALEFLUX WILL BE AT OCP GLOBAL SUMMIT 2019

In a nod to Computational Storage’s increasing traction in the storage industry as a whole, Open Compute Project will host a panel devoted to the technology at its Global Summit held in San Jose. Thad Omura, EVP Marketing of ScaleFlux, will join representatives from six fellow companies to discuss the present state and future development of Computational Storage. The panel will be held on Friday, March 15th, at 4pm in hall 212 AB.

ScaleFlux will additionally be present in SNIA’s booth #A22. The industry standardization authority has made it a priority this year to feature the new Technical Work Group for Computational Storage, for which ScaleFlux is a founding member. ScaleFlux has a private meeting room for face to face discussion with customers, analysts, and partners, and press. (email info@scaleflux.com if interested in hearing about the latest update in Computational Storage)

ScaleFlux™ to Join Computational Storage Panel at OCP Global Summit, March 15
ScaleFlux™ to Join Computational Storage Panel at OCP Global Summit, March 15
industry seeing increasing interest and discussion around moving compute to data industry seeing increasing interest and discussion around moving compute to data
February 8th, 2019 | 2019年2月8日 |
via OCP OCP 网站

SAN JOSE CA -– ScaleFlux is proud to be a part of the Panel Discussion on Computational Storage at Open Compute Project Global Summit!

EVP Marketing Thad Omura will join our fellows hailing from Aclectic Systems, Marvell, Western Digital, Microsoft, NGD Systems, and Eideticom!

This panel will be held Friday, March 15 at 4pm in the 212 AB venue of San Jose Convention Center. See more about OCP here: OCP Site

SAN JOSE CA -– ScaleFlux is proud to be a part of the Panel Discussion on Computational Storage at Open Compute Project Global Summit!

EVP Marketing Thad Omura will join our fellows hailing from Aclectic Systems, Marvell, Western Digital, Microsoft, NGD Systems, and Eideticom!

This panel will be held Friday, March 15 at 4pm in the 212 AB venue of San Jose Convention Center. See more about OCP here: OCP Site

ScaleFlux™ Joins New SNIA Technical Work Group for Computational Storage
ScaleFlux™ Joins New SNIA Technical Work Group for Computational Storage
Industry-wide Interest in Solutions and Standards to Move Compute Closer to Data Industry-wide Interest in Solutions and Standards to Move Compute Closer to Data
November 12th, 2018 | 2018年11月12日 |
via SNIA SNIA 网站

Colorado Springs, CO - The Storage Networking Industry Association (SNIA) today announced the launch of its Computational Storage Technical Work Group (TWG). SNIA member companies intend to come together to create standards to promote the interoperability of computational storage devices, and to define interface standards for system deployment, provisioning, management, and security. This will enable storage architectures and software to be integrated with computation in its many forms.

The industry is seeing an increase in customer requirements to move compute closer to traditional storage devices and systems. In response, a growing number of data-driven applications have demonstrated that adding computation to the normal storage features of devices and systems can realize a significant performance and infrastructure scaling advantage. Computational Storage solutions typically target applications where the demand to process ever-growing storage workloads is outpacing traditional compute server architectures. These applications include AI, big data, content delivery, database, machine learning and many others that are used industry-wide.

“SNIA has an established infrastructure that supports broad member participation and collaboration, producing successful standards and specifications in response to new technological challenges,” said Mark Carlson, SNIA Technical Council Co-Chair. “SNIA is pleased to establish this new Technical Work Group to facilitate the use of Computational Storage in mainstream application environments.”

Computational Storage represents a paradigm shift to heterogenous compute architectures close to storage data workloads which is enabled by standards such as PCIe and NVM Express®. Future low latency interconnect standards such as GenZ and OpenCAPI are also enabling a peer-to-peer type of architecture or even full OS ecosystems for compute, memory, storage and networking, and will also play a role in the evolution of this technology.

“There are a number of applications that can take advantage of computation being moved closer to data, including disaggregation, encryption/decryption, and operating system-capable computation,” said Scott Shadley, SNIA Computational Storage TWG Co-Chair. “The activities of this new TWG will help the industry and users understand and better use this new paradigm of computer architecture.”

Founding members in the Computational Storage TWG include: Arm, Eideticom, Inspur, Lenovo, Micron Technology, Inc., NetApp, NGD Systems, Inc., Nyriad, Samsung Electronics Co. LTD., Scaleflux, SK Hynix, Western Digital Corporation, and Xilinx.

Companies that join SNIA are eligible to participate in this TWG and in 14 other technical work groups. For more information on the Computational Storage Technical Work Group, visit snia.org/computational. For more information on joining SNIA, visit snia.org/become-a-member.

About SNIA

The Storage Networking Industry Association is a not-for-profit global organization, made up of member companies spanning the global storage market. SNIA’s mission is to lead the storage industry worldwide in developing and promoting standards, technologies, and educational services to empower organizations in the management of information. To this end, the SNIA is uniquely committed to delivering standards, education, and services that will propel open storage networking solutions into the broader market. For more information about SNIA, visit https://www.snia.org/.

Colorado Springs, CO - The Storage Networking Industry Association (SNIA) today announced the launch of its Computational Storage Technical Work Group (TWG). SNIA member companies intend to come together to create standards to promote the interoperability of computational storage devices, and to define interface standards for system deployment, provisioning, management, and security. This will enable storage architectures and software to be integrated with computation in its many forms.

The industry is seeing an increase in customer requirements to move compute closer to traditional storage devices and systems. In response, a growing number of data-driven applications have demonstrated that adding computation to the normal storage features of devices and systems can realize a significant performance and infrastructure scaling advantage. Computational Storage solutions typically target applications where the demand to process ever-growing storage workloads is outpacing traditional compute server architectures. These applications include AI, big data, content delivery, database, machine learning and many others that are used industry-wide.

“SNIA has an established infrastructure that supports broad member participation and collaboration, producing successful standards and specifications in response to new technological challenges,” said Mark Carlson, SNIA Technical Council Co-Chair. “SNIA is pleased to establish this new Technical Work Group to facilitate the use of Computational Storage in mainstream application environments.”

Computational Storage represents a paradigm shift to heterogenous compute architectures close to storage data workloads which is enabled by standards such as PCIe and NVM Express®. Future low latency interconnect standards such as GenZ and OpenCAPI are also enabling a peer-to-peer type of architecture or even full OS ecosystems for compute, memory, storage and networking, and will also play a role in the evolution of this technology.

“There are a number of applications that can take advantage of computation being moved closer to data, including disaggregation, encryption/decryption, and operating system-capable computation,” said Scott Shadley, SNIA Computational Storage TWG Co-Chair. “The activities of this new TWG will help the industry and users understand and better use this new paradigm of computer architecture.”

Founding members in the Computational Storage TWG include: Arm, Eideticom, Inspur, Lenovo, Micron Technology, Inc., NetApp, NGD Systems, Inc., Nyriad, Samsung Electronics Co. LTD., Scaleflux, SK Hynix, Western Digital Corporation, and Xilinx.

Companies that join SNIA are eligible to participate in this TWG and in 14 other technical work groups. For more information on the Computational Storage Technical Work Group, visit snia.org/computational. For more information on joining SNIA, visit snia.org/become-a-member.

About SNIA

The Storage Networking Industry Association is a not-for-profit global organization, made up of member companies spanning the global storage market. SNIA’s mission is to lead the storage industry worldwide in developing and promoting standards, technologies, and educational services to empower organizations in the management of information. To this end, the SNIA is uniquely committed to delivering standards, education, and services that will propel open storage networking solutions into the broader market. For more information about SNIA, visit https://www.snia.org/.

ScaleFlux™ is a Silver Sponsor at Three Upcoming Xilinx Conferences
ScaleFlux™ is a Silver Sponsor at Three Upcoming Xilinx Conferences
exhibiting at Xilinx Developer Forum San Jose, Beijing, and Frankfurt exhibiting at Xilinx Developer Forum San Jose, Beijing, and Frankfurt
September 27th, 2018 2018年9月27日

SAN JOSE CA -– ScaleFlux is proud to be exhibiting at three Xilinx Developer Conferences on three different continents!

ScaleFlux Computational Storage Xilinx Developer Forum San Jose

October 2nd | XDF San Jose: Thad Omura, our EVP of marketing, will be speaking at 2:15pm in the Fairmont San Jose. We will also be running live Computational Storage demos at booth 14! See details.

ScaleFlux Computational Storage Xilinx Developer Forum Beijing

October 16th | XDF Beijing: Come visit our exhibition table and meet our Beijing/Shanghai team members! See details.

ScaleFlux Computational Storage Xilinx Developer Forum Frankfurt

December 10th | XDF Frankfurt: Currently in planning! We have some exciting new developments to unveil in December. See details.

Hope you join us at these exciting events! #compute2data

SAN JOSE CA -– ScaleFlux is proud to be exhibiting at three Xilinx Developer Conferences on three different continents!

ScaleFlux Computational Storage Xilinx Developer Forum San Jose

October 2nd | XDF San Jose: Thad Omura, our EVP of marketing, will be speaking at 2:15pm in the Fairmont San Jose. We will also be running live Computational Storage demos at booth 14! See details.

ScaleFlux Computational Storage Xilinx Developer Forum Beijing

October 16th | XDF Beijing: Come visit our exhibition table and meet our Beijing/Shanghai team members! See details.

ScaleFlux Computational Storage Xilinx Developer Forum Frankfurt

December 10th | XDF Frankfurt: Currently in planning! We have some exciting new developments to unveil in December. See details.

Hope you join us at these exciting events! #compute2data

ScaleFlux™ Wins Best-of-Show Award at Flash Memory Summit 2018
ScaleFlux™ Wins Best-of-Show Award at Flash Memory Summit 2018
Major Conference Presence Capped by Recognition as Most Innovative Startup Major Conference Presence Capped by Recognition as Most Innovative Startup
August 10th, 2018 | 2018年8月10日 |
via BusinessWire BusinessWire 网站

SAN JOSE, Calif. -- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, is proud to receive a Flash Memory Summit best-of-show award during a remarkable debutante appearance at the industry conference. ScaleFlux’s win of the Most Innovative Flash Memory Startup category is an exciting capstone to a busy week that included a well-trafficked expo booth, recognition as the market leader for Computational Storage by IT Brand Pulse, and a rousing keynote presentation. The ScaleFlux staff are humbled and thrilled by the show response and will build on the momentum to bring the CSS solution to mass market adoption.

“Computational Storage is the next big thing to address the performance needs of customers implementing big data applications at scale,” said Jay Kramer, Chairman of the Awards Program and President of Network Storage Advisors Inc. “We are proud to recognize ScaleFlux with their CSS 1000 Series achieving incredible transaction-per-second throughput by bringing heavy duty compute engines to the data to address predictable QoS performance for business-critical applications.”

ScaleFlux CEO and Co-Founder Hao Zhong’s Wednesday afternoon keynote announced expanding hyperscale usage, delved into the theory behind the company’s development process, and delivered a compelling vision for Computational Storage at large.

“It is my pleasure and privilege to accept the award for Most Innovative Startup on behalf of the ScaleFlux team,” said Hao. “Being selected by Flash Memory Summit among industry greats lends credibility to the traction of our product and direction. Look to us for continued innovation as we bring Computational Storage to the mainstream market.”

ScaleFlux continues to refine its key technology with targets on widespread usage. The newly released 2nd generation platform brings higher performance, lower latency, expanded capacity capability, support for leading-edge 3D TLC NAND, customizable database engine accelerators, and more. In addition, cloud access to Computational Storage via a partnership with Packet allows the easiest accessibility yet at low opportunity cost.

SAN JOSE, Calif. -- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, is proud to receive a Flash Memory Summit best-of-show award during a remarkable debutante appearance at the industry conference. ScaleFlux’s win of the Most Innovative Flash Memory Startup category is an exciting capstone to a busy week that included a well-trafficked expo booth, recognition as the market leader for Computational Storage by IT Brand Pulse, and a rousing keynote presentation. The ScaleFlux staff are humbled and thrilled by the show response and will build on the momentum to bring the CSS solution to mass market adoption.

“Computational Storage is the next big thing to address the performance needs of customers implementing big data applications at scale,” said Jay Kramer, Chairman of the Awards Program and President of Network Storage Advisors Inc. “We are proud to recognize ScaleFlux with their CSS 1000 Series achieving incredible transaction-per-second throughput by bringing heavy duty compute engines to the data to address predictable QoS performance for business-critical applications.”

ScaleFlux CEO and Co-Founder Hao Zhong’s Wednesday afternoon keynote announced expanding hyperscale usage, delved into the theory behind the company’s development process, and delivered a compelling vision for Computational Storage at large.

“It is my pleasure and privilege to accept the award for Most Innovative Startup on behalf of the ScaleFlux team,” said Hao. “Being selected by Flash Memory Summit among industry greats lends credibility to the traction of our product and direction. Look to us for continued innovation as we bring Computational Storage to the mainstream market.”

ScaleFlux continues to refine its key technology with targets on widespread usage. The newly released 2nd generation platform brings higher performance, lower latency, expanded capacity capability, support for leading-edge 3D TLC NAND, customizable database engine accelerators, and more. In addition, cloud access to Computational Storage via a partnership with Packet allows the easiest accessibility yet at low opportunity cost.

ScaleFlux™ Computational Storage Transforms Cloud Infrastructure
ScaleFlux™ Computational Storage Transforms Cloud Infrastructure
Market-wide expansion highlighted by Hyperscale usage at Flash Memory Summit 2018 Market-wide expansion highlighted by Hyperscale usage at Flash Memory Summit 2018
August 6th, 2018 | 2018年8月6日 |
via BusinessWire BusinessWire 网站

SANTA CLARA, Calif. -- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, is rapidly expanding its market reach directly with Tier-1 customers worldwide. The launch of its 2nd generation Computational Storage platform brings support for cutting edge NAND technology and delivers performance enhancements across the board to maintain its lead in maximizing flash infrastructure.

Hao Zhong, CEO and Co-Founder of ScaleFlux, will deliver a keynote presentation titled “Computational Storage: Acceleration Through Intelligence and Agility”, Wednesday, Aug 8 @ 2:10pm in the main hall of Flash Memory Summit. This presentation will cover how hyperscale/cloud environments are realizing a 10X improvement in hybrid transactional and analytical processing capabilities. In addition, webscalers in markets spanning online gaming, adtech, and digital payments who heavily depend on high transaction applications have realized the benefits of ScaleFlux Computational Storage over other Flash storage solutions.

“PhonePe is leading India’s digital payments revolution through super reliable payment processing infrastructure that continues to grow our transactions threefold year on year,” said Burzin Engineer, Chief Reliability Officer for PhonePe, a Flipkart Company. “ScaleFlux is accelerating multiple, business-critical applications for us so we can most efficiently scale our low-latency, Flash storage deployment.”

2nd GENERATION PLATFORM LAUNCHED

ScaleFlux has released its 2nd generation product platform highlighted by several key new capabilities and features:

Rev 2.x of its host-controlled Flash translation and management software solution that provides higher throughput, lower latency, VMware ESXi 6.5 & 6.5 U1 support, and expanded storage capacity support Support for industry leading, state-of-the-art 3D TLC NAND technologies from both Toshiba and Micron Customizable database engine accelerator for transactional and analytical workloads delivering an order-of-magnitude performance improvement vs. NVMe SSDs “We are delighted to expand Computational Storage to mainstream infrastructure deployment,” said Thad Omura, EVP Marketing for ScaleFlux. “We integrate intelligence and agility from the application down through the Flash controller and compute engine hardware. This has enabled our customers to quickly bring both custom and turnkey applications to market with substantial value over legacy SSD solutions.”

COME VISIT THE COMPUTATIONAL STORAGE MARKET LEADER @ FMS 2018

ScaleFlux is a premier sponsor of Flash Memory Summit 2018, held at the Santa Clara Convention Center on August 7-9. The ScaleFlux executive team and engineering/business staff will be present to discuss the latest market applications and product solutions in Computational Storage highlighted by the following:

  • Jam-packed expo display: Booth #113 during exhibition hours with live demonstrations and exciting giveaways; Inspur Systems will be at the booth to demonstrate turnkey Hadoop 3.0 accelerated server and storage solutions
  • Business session: Iqbal Bhalwani (VP WW Sales of ScaleFlux) will be on Panel Session BMKT-2012B-1: “What do the 800-Pound Gorilla Customers Want In Flash Storage?”, Tuesday, Aug 7 @4:55pm as part of the Business/Marketing Track
  • Reception of a distinguished industry award: In June, IT Brand Pulse awarded ScaleFlux as the Market Leader of Computational Storage (in terms of price, performance, reliability, service and innovation) in the field of Flash Storage Brand Leaders; come help us celebrate the first award of this new category Wednesday, Aug 8 @ 2:00pm in the main hall
  • Conference keynote presentation: Hao Zhong (CEO and Co-Founder of ScaleFlux) will be onstage to deliver: “Computational Storage: Acceleration Through Intelligence and Agility”, Wednesday, Aug 8 @ 2:10pm in the main hall
  • Technical session: Thad Omura (EVP Marketing of ScaleFlux) will present at Technical Session COMP-301-1: “Increasing Performance by Moving Compute Closer to Data”, Thursday, Aug 9 @ 8:30am as part of the Computational Storage Track
  • Private meeting room: On-site scheduled meetings with customers, partners, press and analysts (email info@scaleflux.com if interested)
  • SANTA CLARA, Calif. -- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, is rapidly expanding its market reach directly with Tier-1 customers worldwide. The launch of its 2nd generation Computational Storage platform brings support for cutting edge NAND technology and delivers performance enhancements across the board to maintain its lead in maximizing flash infrastructure.

    Hao Zhong, CEO and Co-Founder of ScaleFlux, will deliver a keynote presentation titled “Computational Storage: Acceleration Through Intelligence and Agility”, Wednesday, Aug 8 @ 2:10pm in the main hall of Flash Memory Summit. This presentation will cover how hyperscale/cloud environments are realizing a 10X improvement in hybrid transactional and analytical processing capabilities. In addition, webscalers in markets spanning online gaming, adtech, and digital payments who heavily depend on high transaction applications have realized the benefits of ScaleFlux Computational Storage over other Flash storage solutions.

    “PhonePe is leading India’s digital payments revolution through super reliable payment processing infrastructure that continues to grow our transactions threefold year on year,” said Burzin Engineer, Chief Reliability Officer for PhonePe, a Flipkart Company. “ScaleFlux is accelerating multiple, business-critical applications for us so we can most efficiently scale our low-latency, Flash storage deployment.”

    2nd GENERATION PLATFORM LAUNCHED

    ScaleFlux has released its 2nd generation product platform highlighted by several key new capabilities and features:

    Rev 2.x of its host-controlled Flash translation and management software solution that provides higher throughput, lower latency, VMware ESXi 6.5 & 6.5 U1 support, and expanded storage capacity support Support for industry leading, state-of-the-art 3D TLC NAND technologies from both Toshiba and Micron Customizable database engine accelerator for transactional and analytical workloads delivering an order-of-magnitude performance improvement vs. NVMe SSDs “We are delighted to expand Computational Storage to mainstream infrastructure deployment,” said Thad Omura, EVP Marketing for ScaleFlux. “We integrate intelligence and agility from the application down through the Flash controller and compute engine hardware. This has enabled our customers to quickly bring both custom and turnkey applications to market with substantial value over legacy SSD solutions.”

    COME VISIT THE COMPUTATIONAL STORAGE MARKET LEADER @ FMS 2018

    ScaleFlux is a premier sponsor of Flash Memory Summit 2018, held at the Santa Clara Convention Center on August 7-9. The ScaleFlux executive team and engineering/business staff will be present to discuss the latest market applications and product solutions in Computational Storage highlighted by the following:

    • Jam-packed expo display: Booth #113 during exhibition hours with live demonstrations and exciting giveaways; Inspur Systems will be at the booth to demonstrate turnkey Hadoop 3.0 accelerated server and storage solutions
    • Business session: Iqbal Bhalwani (VP WW Sales of ScaleFlux) will be on Panel Session BMKT-2012B-1: “What do the 800-Pound Gorilla Customers Want In Flash Storage?”, Tuesday, Aug 7 @4:55pm as part of the Business/Marketing Track
    • Reception of a distinguished industry award: In June, IT Brand Pulse awarded ScaleFlux as the Market Leader of Computational Storage (in terms of price, performance, reliability, service and innovation) in the field of Flash Storage Brand Leaders; come help us celebrate the first award of this new category Wednesday, Aug 8 @ 2:00pm in the main hall
    • Conference keynote presentation: Hao Zhong (CEO and Co-Founder of ScaleFlux) will be onstage to deliver: “Computational Storage: Acceleration Through Intelligence and Agility”, Wednesday, Aug 8 @ 2:10pm in the main hall
    • Technical session: Thad Omura (EVP Marketing of ScaleFlux) will present at Technical Session COMP-301-1: “Increasing Performance by Moving Compute Closer to Data”, Thursday, Aug 9 @ 8:30am as part of the Computational Storage Track
    • Private meeting room: On-site scheduled meetings with customers, partners, press and analysts (email info@scaleflux.com if interested)
ScaleFlux™ and Packet First to Provide Computational Storage in the Cloud
ScaleFlux™ and Packet First to Provide Computational Storage in the Cloud
Bare metal cloud instances supercharged for data-driven applications Bare metal cloud instances supercharged for data-driven applications
August 3rd, 2018 | 2018年8月3日 |
via BusinessWire BusinessWire 网站

SAN JOSE, CA -– ScaleFlux, Inc., the pioneer in Computational Storage, announced today that its CSS 1000 solution is now available through Packet, the leading bare metal cloud. With access to ScaleFlux Computational Storage in select Packet configurations, customers can deploy in minutes ScaleFlux’s unique flash technology, dramatically improving performance for transactional and analytical database applications.

“As deployment ramps for ScaleFlux Computational Storage, our customers are demanding the flexibility to consume our product both on-prem and in the cloud,” said Iqbal Bhalwani, VP of Global Sales for ScaleFlux. “With their global reach, strong enterprise experience and flexible deployment model, we’re delighted to partner with Packet to extend the accessibility of our transformative Computational Storage technology.”

Packet, which was founded in 2014 and is backed by SoftBank, provides automated access to state-of-the-art physical server infrastructure. The first deployment of ScaleFlux Computational Storage is based on Packet’s m2.xlarge configuration with dual Xeon Gold 5120s, 384GB DRAM, and two ScaleFlux CSS 1000 U.2 3.2TB drives. Any Linux-based application can immediately take advantage of the CSS 1000 low-latency Flash storage. Popular database and big data applications can further benefit from hardware-accelerated compression and erasure coding.

“We’ve created an easy-to-use cloud infrastructure experience for innovators that are looking to get top application level performance by leveraging cutting-edge hardware-accelerated solutions,” said Jacob Smith, SVP of Engagement and a co-founder at Packet. “Enabling developer-style access to ScaleFlux’s Computational Storage solution fits right into this theme and we’re excited to be the first cloud provider to offer this technology.”

CHECK OUT COMPUTATIONAL STORAGE ON THE CLOUD @ FMS 2018, BOOTH #113

ScaleFlux is a premier sponsor of Flash Memory Summit, held at the Santa Clara Convention Center on August 7-9. Come visit us in Booth #113 during showcase hours and see a NoSQL database benchmark running real-time on Packet infrastructure. For information about accessing ScaleFlux Computational Storage on Packet, please inquire at info@scaleflux.com.

ScaleFlux Computational Storage Cloud Integration Packet

SAN JOSE, CA -– ScaleFlux, Inc., the pioneer in Computational Storage, announced today that its CSS 1000 solution is now available through Packet, the leading bare metal cloud. With access to ScaleFlux Computational Storage in select Packet configurations, customers can deploy in minutes ScaleFlux’s unique flash technology, dramatically improving performance for transactional and analytical database applications.

“As deployment ramps for ScaleFlux Computational Storage, our customers are demanding the flexibility to consume our product both on-prem and in the cloud,” said Iqbal Bhalwani, VP of Global Sales for ScaleFlux. “With their global reach, strong enterprise experience and flexible deployment model, we’re delighted to partner with Packet to extend the accessibility of our transformative Computational Storage technology.”

Packet, which was founded in 2014 and is backed by SoftBank, provides automated access to state-of-the-art physical server infrastructure. The first deployment of ScaleFlux Computational Storage is based on Packet’s m2.xlarge configuration with dual Xeon Gold 5120s, 384GB DRAM, and two ScaleFlux CSS 1000 U.2 3.2TB drives. Any Linux-based application can immediately take advantage of the CSS 1000 low-latency Flash storage. Popular database and big data applications can further benefit from hardware-accelerated compression and erasure coding.

“We’ve created an easy-to-use cloud infrastructure experience for innovators that are looking to get top application level performance by leveraging cutting-edge hardware-accelerated solutions,” said Jacob Smith, SVP of Engagement and a co-founder at Packet. “Enabling developer-style access to ScaleFlux’s Computational Storage solution fits right into this theme and we’re excited to be the first cloud provider to offer this technology.”

CHECK OUT COMPUTATIONAL STORAGE ON THE CLOUD @ FMS 2018, BOOTH #113

ScaleFlux is a premier sponsor of Flash Memory Summit, held at the Santa Clara Convention Center on August 7-9. Come visit us in Booth #113 during showcase hours and see a NoSQL database benchmark running real-time on Packet infrastructure. For information about accessing ScaleFlux Computational Storage on Packet, please inquire at info@scaleflux.com.

ScaleFlux Computational Storage Cloud Integration Packet
ScaleFlux Showcasing CSS at Two Events in July! ScaleFlux Showcasing CSS at Two Events in July!
speaking series drive worldwide outreach speaking series drive worldwide outreach
June 29th, 2018 2018年6月29日

SAN JOSE CA -– ScaleFlux is proud to be speaking at two great international shows in July!

July 5th | Boston Technology Innovation Day 2018: Daniel Schmidt, our Director of Business Development in EMEA, will present "Unleashing Data Driven Applications at Scale" in the company of a whole host of industry movers & shakers in Germany! See details.

July 20th | Global Storage Semicon: Hao Zhong, our CEO and Co-Founder, will speak at this global flash memory summit held in Wuhan, China. This packed event attracts a wide array of Chinese and international talent as data storage and processing demands reach an all-time high. See details.

Hope you join us at these exciting events! #compute2data

ScaleFlux Computational Storage Speaking Events

SAN JOSE CA -– ScaleFlux is proud to be speaking at two great international shows in July!

July 5th | Boston Technology Innovation Day 2018: Daniel Schmidt, our Director of Business Development in EMEA, will present "Unleashing Data Driven Applications at Scale" in the company of a whole host of industry movers & shakers in Germany! See details.

July 20th | Global Storage Semicon: Hao Zhong, our CEO and Co-Founder, will speak at this global flash memory summit held in Wuhan, China. This packed event attracts a wide array of Chinese and international talent as data storage and processing demands reach an all-time high. See details.

Hope you join us at these exciting events! #compute2data

ScaleFlux Computational Storage Speaking Events
ScaleFlux Raises $25M Series B Funding
to Expand Global Business and Fuel Innovation Growth
ScaleFlux Raises $25M Series B Funding
to Expand Global Business and Fuel Innovation Growth
Production Computational Storage Deployments Accelerating Data Driven Applications Industry-Wide Production Computational Storage Deployments Accelerating Data Driven Applications Industry-Wide
March 20th, 2018 | 2018年3月20日 |
via BusinessWire BusinessWire 网站

SAN JOSE, CA -– ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, announced it has raised $25M in Series B funding to accelerate global expansion and new product innovation. The round was led by Shunwei Capital and includes multiple Tier-1 strategic corporate investors. This round brings the total invested to date over $40M.

“ScaleFlux Computational Storage is poised to revolutionize how the most demanding transactional and analytical data center applications scale and generate revenue for businesses industry-wide. Easy integration into standard server and storage systems makes ScaleFlux Computational Storage an incredible asset to Cloud and Webscale data center environments,” said Tuck Lye Koh, Founding Partner and CEO of Shunwei Capital. “We are thrilled to work with the ScaleFlux team who has efficiently delivered high-value innovation into the multi-billion dollar flash storage market, with the agility and execution capabilities to quickly support next generation applications as they emerge in this vibrant data-driven ecosystem.”

Shipping in volume for revenue production, ScaleFlux Computational Storage is proven across Database, Big Data, and Data Warehousing applications to dramatically improve IT infrastructure efficiency and enable customers with new revenue generating application performance. As easy to deploy as a standard PCIe SSD, ScaleFlux Computational Storage extends low-latency 3D-NAND Flash storage to address both I/O and algorithmically intense computational bottlenecks. By cohesively solving storage and compute bottlenecks together, application run-time is minimized, flash capacity utilization is optimized, and total cost of ownership is enhanced for advanced data center compute and storage infrastructure.

“It is an honor to add Shunwei Capital to our Board of Directors and to directly benefit from their extensive knowledge, experience, and industry network to quickly scale our business across the entire IT internet industry,” said Hao Zhong, Co-Founder and CEO of ScaleFlux. “This funding will accelerate our Computational Storage deployments in global markets while also expanding the number of applications and use cases we support.”

About Shunwei Capital

Shunwei Capital was founded in 2011 by Jun Lei (also the founder of Xiaomi) and Tuck Lye Koh with over 2 billion US dollars in funds under management. Shunwei Capital focuses on early and growth stage startups that are poised to fundamentally disrupt how businesses and services are conducted in traditional industry verticals with differentiated Internet and IoT technology.

SAN JOSE CA, March 20, 2018 – ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, announced it has raised $25M in Series B funding to accelerate global expansion and new product innovation. The round was led by Shunwei Capital and includes multiple Tier-1 strategic corporate investors. This round brings the total invested to date over $40M.

“ScaleFlux Computational Storage is poised to revolutionize how the most demanding transactional and analytical data center applications scale and generate revenue for businesses industry-wide. Easy integration into standard server and storage systems makes ScaleFlux Computational Storage an incredible asset to Cloud and Webscale data center environments,” said Tuck Lye Koh, Founding Partner and CEO of Shunwei Capital. “We are thrilled to work with the ScaleFlux team who has efficiently delivered high-value innovation into the multi-billion dollar flash storage market, with the agility and execution capabilities to quickly support next generation applications as they emerge in this vibrant data-driven ecosystem.”

Shipping in volume for revenue production, ScaleFlux Computational Storage is proven across Database, Big Data, and Data Warehousing applications to dramatically improve IT infrastructure efficiency and enable customers with new revenue generating application performance. As easy to deploy as a standard PCIe SSD, ScaleFlux Computational Storage extends low-latency 3D-NAND Flash storage to address both I/O and algorithmically intense computational bottlenecks. By cohesively solving storage and compute bottlenecks together, application run-time is minimized, flash capacity utilization is optimized, and total cost of ownership is enhanced for advanced data center compute and storage infrastructure.

“It is an honor to add Shunwei Capital to our Board of Directors and to directly benefit from their extensive knowledge, experience, and industry network to quickly scale our business across the entire IT internet industry,” said Hao Zhong, Co-Founder and CEO of ScaleFlux. “This funding will accelerate our Computational Storage deployments in global markets while also expanding the number of applications and use cases we support.”

About Shunwei Capital

Shunwei Capital was founded in 2011 by Jun Lei (also the founder of Xiaomi) and Tuck Lye Koh with over 2 billion US dollars in funds under management. Shunwei Capital focuses on early and growth stage startups that are poised to fundamentally disrupt how businesses and services are conducted in traditional industry verticals with differentiated Internet and IoT technology.

GlobalDots Feature: CSS Product Brief (Russian) GlobalDots Feature: CSS Product Brief (Russian)
March 19th, 2018 | 2018年3月19日 |
Via Habrahabr.ru Habrahabr.ru 网站

Our worldwide partner GlobalDots wrote a post about the CSS product line and accelerated applications on the Russian blog Habrahabr.ru, a hub frequented by tech industry veterans from across many disciplines. This product brief serves as an excellent point of exposure for Computational Storage.

Read the post (in Russian) here!

Our worldwide partner GlobalDots wrote a post about the CSS product line and accelerated applications on the Russian blog Habrahabr.ru, a hub frequented by tech industry veterans from across many disciplines. This product brief serves as an excellent point of exposure for Computational Storage.

Read the post (in Russian) here!

ScaleFlux and GlobalDots Partner for CloudFest Exhibition ScaleFlux and GlobalDots Partner for CloudFest Exhibition
March 12th, 2018 | 2018年3月12日 |
View Conference Agenda 会议日程
ScaleFlux Computational Storage Cloudfest GlobalDots

EUROPA-PARK, Germany -– Hot on the heels of our partnership announcement, ScaleFlux and GlobalDots will jointly exhibit at CloudFest this week in Germany! Come by Booth R16 to see our combined effort in bringing Computational Storage acceleration to data-driven application customers worldwide!

EUROPA PARK, Germany -– Hot on the heels of our partnership announcement, ScaleFlux and GlobalDots will jointly exhibit at CloudFest this week in Germany! Come by Booth R16 to see our combined effort in bringing Computational Storage acceleration to data-driven application customers worldwide!

Partner Meetup in Shanghai with Aerospike and CTrip Partner Meetup in Shanghai with Aerospike and CTrip
March 10th, 2018 | 2018年3月10日 |
View Event Agenda 信息
ScaleFlux Computational Storage Aeropsike Acceleration

SHANGHAI, China -– ScaleFlux is proud to sponsor a meetup in Shanghai in partnership with Aerospike and Ctrip. The event will highlight the innovation of Computational Storage in practical customer use cases spanning global Internet advertising, finance, payment, and risk control platforms.

Held at Ctrip headquarters in Shanghai, this exciting event draws an audience of professionals at the industry forefront. Featured speakers include Dr. Srini V. Srinivasan of Aerospike, Dr. Zhong Hao of ScaleFlux, Mr. Chen Dong of 360商业产品, and Mr. Bruce Li of Paypal. Please see the full event agenda for talk details.

SHANGHAI, China -– ScaleFlux is proud to sponsor a meetup in Shanghai in partnership with Aerospike and Ctrip. The event will highlight the innovation of Computational Storage in practical customer use cases spanning global Internet advertising, finance, payment, and risk control platforms.

Held at Ctrip headquarters in Shanghai, this exciting event draws an audience of professionals at the industry forefront. Featured speakers include Dr. Srini V. Srinivasan of Aerospike, Dr. Zhong Hao of ScaleFlux, Mr. Chen Dong of 360商业产品, and Mr. Bruce Li of Paypal. Please see the full event agenda for talk details.

GlobalDots and ScaleFlux Announce the Solid State Drive (SSD) Revolution in Database Computing GlobalDots and ScaleFlux Announce the Solid State Drive (SSD) Revolution in Database Computing
March 6th, 2018 | 2018年3月6日 |
via GlobalDots 通过 GlobalDots
ScaleFlux Computational Storage GlobalDots Partnership

BERLIN, Germany -- GlobalDots, a leading web performance and cloud security system integrator announced today a partnership with ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, to deliver products that greatly optimize infrastructure efficiency and scaling for hyperscale and webscale clients worldwide.

ScaleFlux ships the industry’s first production qualified Computational Storage Subsystem (CSS) products for customer deployment. CSS alleviates expensive CPU processing limitations by bringing computationally advanced algorithms straight to the data, simultaneously defeating storage and compute bottlenecks. The combination of terabytes of 3D NAND flash, reprogrammable low-power FPGAs, and state-of-the-art flash controller technology enables powerful validated acceleration for an increasing list of data-driven applications:

Efficiency Advantage of CSS vs. Standard NVMe:

  • 2.7x better transaction rate with Aerospike
  • 3.5x better transaction rate with PostgreSQL
  • 3.6x better flash capacity utilization with MySQL
  • 2.3x faster queries with Vitesse Data Deepgreen Db
  • 2.2x better KV-Store write throughput with RocksDB
  • 2.4x better transaction rate with Apache HBase
  • 1.7x more jobs completed (vs. HDD) with Hadoop
  • 1.7x more jobs completed with Apache Spark

CSS products have been shipping to customers worldwide to proven success. Easily installed into servers through standard PCIe HHHL AIC and 2.5” U.2 drive level form factors, CSS creates immediate impact in datacenter scaling capability and cost reduction. ScaleFlux’s strategic partnership with GlobalDots will carry this momentum to the waiting hands of a widespread audience.

“It is my pleasure to announce that GlobalDots is now a proud partner of ScaleFlux. ScaleFlux has an amazing product that will dramatically change how we approach database computing, and this partnership offers us a way to bring them closer to more satisfied end-users.” said Thorsten Deutrich, VP Sales of GlobalDots.

What makes GlobalDots so unique is the fact that they work with practically every vendor in the industry and specific verticals. This enables their team to have an in-depth knowledge of every specific technology and solution – knowledge, real-world scenarios and use cases that internal IT teams often lack due to complexity and time needed to learn, analyze, test, deploy and monitor in production environments.

“GlobalDots will accelerate the proliferation of Computational Storage with its worldwide customer base whose business is dependent on data-driven applications. We are looking forward to partnering as we jointly deliver innovative solutions that change the way both compute and storage infrastructure scales and enable new revenue generation opportunities for our customers,” said Thad Omura, EVP Marketing for ScaleFlux.

GlobalDots and ScaleFlux will present Computational Storage Subsystem 1000 Series at the upcoming CloudFest in Rust, Germany from March 10-16th.

About GlobalDots (www.globaldots.com)

GlobalDots is a leading web performance and cloud security system integrator, hardware broker and colocation provider with over 15 years of experience. GlobalDots’ unique positioning on the global market enables them to be at the forefront of every new technology and with that unique expertise GlobalDots helps numerous clients to quickly and cost effectively choose, test and deploy the best solutions to service their customer base.

Their clients range from Fortune 500 to small & medium enterprise to promising startups in practically every industry and vertical.

Iconic brands like Lufthansa, Playtika, Trading View, Lamborghini, Bosch, Fiat, Rocket Internet, Benetton and Bulova trust GlobalDots to help them accelerate and secure their digital assets.

BERLIN, Germany -- GlobalDots, a leading web performance and cloud security system integrator announced today a partnership with ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, to deliver products that greatly optimize infrastructure efficiency and scaling for hyperscale and webscale clients worldwide.

ScaleFlux ships the industry’s first production qualified Computational Storage Subsystem (CSS) products for customer deployment. CSS alleviates expensive CPU processing limitations by bringing computationally advanced algorithms straight to the data, simultaneously defeating storage and compute bottlenecks. The combination of terabytes of 3D NAND flash, reprogrammable low-power FPGAs, and state-of-the-art flash controller technology enables powerful validated acceleration for an increasing list of data-driven applications:

Efficiency Advantage of CSS vs. Standard NVMe:

  • 2.7x better transaction rate with Aerospike
  • 3.5x better transaction rate with PostgreSQL
  • 3.6x better flash capacity utilization with MySQL
  • 2.3x faster queries with Vitesse Data Deepgreen Db
  • 2.2x better KV-Store write throughput with RocksDB
  • 2.4x better transaction rate with Apache HBase
  • 1.7x more jobs completed (vs. HDD) with Hadoop
  • 1.7x more jobs completed with Apache Spark

CSS products have been shipping to customers worldwide to proven success. Easily installed into servers through standard PCIe HHHL AIC and 2.5” U.2 drive level form factors, CSS creates immediate impact in datacenter scaling capability and cost reduction. ScaleFlux’s strategic partnership with GlobalDots will carry this momentum to the waiting hands of a widespread audience.

“It is my pleasure to announce that GlobalDots is now a proud partner of ScaleFlux. ScaleFlux has an amazing product that will dramatically change how we approach database computing, and this partnership offers us a way to bring them closer to more satisfied end-users.” said Thorsten Deutrich, VP Sales of GlobalDots.

What makes GlobalDots so unique is the fact that they work with practically every vendor in the industry and specific verticals. This enables their team to have an in-depth knowledge of every specific technology and solution – knowledge, real-world scenarios and use cases that internal IT teams often lack due to complexity and time needed to learn, analyze, test, deploy and monitor in production environments.

“GlobalDots will accelerate the proliferation of Computational Storage with its worldwide customer base whose business is dependent on data-driven applications. We are looking forward to partnering as we jointly deliver innovative solutions that change the way both compute and storage infrastructure scales and enable new revenue generation opportunities for our customers,” said Thad Omura, EVP Marketing for ScaleFlux.

GlobalDots and ScaleFlux will present Computational Storage Subsystem 1000 Series at the upcoming CloudFest in Rust, Germany from March 10-16th.

About GlobalDots (www.globaldots.com)

GlobalDots is a leading web performance and cloud security system integrator, hardware broker and colocation provider with over 15 years of experience. GlobalDots’ unique positioning on the global market enables them to be at the forefront of every new technology and with that unique expertise GlobalDots helps numerous clients to quickly and cost effectively choose, test and deploy the best solutions to service their customer base.

Their clients range from Fortune 500 to small & medium enterprise to promising startups in practically every industry and vertical.

Iconic brands like Lufthansa, Playtika, Trading View, Lamborghini, Bosch, Fiat, Rocket Internet, Benetton and Bulova trust GlobalDots to help them accelerate and secure their digital assets.

ScaleFlux First to Revenue Production With Computational Storage ScaleFlux First to Revenue Production With Computational Storage
Transformative Data-Driven Technology Showcased at Gartner IT Data Center Conference Transformative Data-Driven Technology Showcased at Gartner IT Data Center Conference
December 4th, 2017 | 2017年12月4日 |
via Business Wire 通过 Business Wire
ScaleFlux Computational Storage Gartner Las Vegas

LAS VEGAS -- ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, announced today it is shipping its Computational Storage Subsystem (CSS) for revenue into multiple enterprise end-user production environments. Uniquely architected to address the plethora of quickly evolving Data-Driven applications, CSS is proven across Database, Big Data, Data Warehousing and other use cases to dramatically improve IT infrastructure efficiency. By extending low-latency PCIe 3D-NAND SSDs to address both storage I/O and algorithmically intense computational bottlenecks, CSS uniquely minimizes application run-time, optimized flash capacity utilization, and reduced total cost of ownership for advanced data center compute and storage infrastructure.

“From high transaction throughput and e-commerce payment environments to travel sites that demand real time response to queries, our customers are getting more useful work from their volume flash storage deployments with CSS,” said Hao Zhong, Co-Founder and CEO of ScaleFlux. “Leveraging programmable hardware, we are able to address new customer demands on a multi-month cycle versus a multi-year ASIC development cycle for new intelligent storage features, which puts us in an advantageous position in this fast-moving market.”

CSS easily integrates into all standard Linux / x86 server flash storage environments and has been validated to improve the efficiency of popular applications including Aerospike, PostgreSQL, MySQL, DeepgreenDB, RocksDB, HBase, Hadoop, and Spark (see Application Value Page for benchmark details). The list of applications continues to expand as ScaleFlux innovates from both a compute engine and Storage I/O acceleration perspective, and will extend into content delivery, search, HPC, AI and machine learning environments.

“ScaleFlux CSS has proven to deliver extremely high transactions with consistently low latency – something that is becoming more of a challenge as 3D NAND evolves to larger and larger die densities,” said Brian Bulkowski, Co-Founder & CTO for Aerospike. “We are excited to collaborate with ScaleFlux on future acceleration functions that can extend Aerospike’s value proposition as the leading, large scale Hybrid Memory Database for real-time applications.”

ScaleFlux CSS is available today in both PCIe Card and U.2 form factors with up to 6.4 Terabytes of state-of-the-art 3D NAND. Through a turnkey, easily-to-install software package, both low-latency storage I/O and compute hardware acceleration are easily enabled without development effort or application re-compilation.

“Computational Storage is an innovative solution to complex, application-level bottlenecks,” said Jim Handy, General Director of Objective Analysis. “ScaleFlux has found a way to combine this concept with the dramatic increase in flash deployment in virtually every data center.”

Come Visit ScaleFlux at the Gartner IT Data Center Conference -- Venetian Hotel, Las Vegas, NV, December 4-7, Solution Showcase Hall C Level 2, Inspur Booth #351

Industry experts and end users will come together to discuss and evaluate the latest data center IT technologies at the Gartner IT Data Center Conference. ScaleFlux will be running a live demonstration of an Aerospike NoSQL Database accelerated by low-latency CSS on an Inspur 1U, volume server. Inspur, a leading data center and cloud computing total solutions provider and worldwide top 4 largest server manufacturer, has fully integrated, tested, and validated CSS on its platform and has declared ScaleFlux a certified partner.

LAS VEGAS -- ScaleFlux, Inc., the pioneer in the deployment of Computational Storage at scale, announced today it is shipping its Computational Storage Subsystem (CSS) for revenue into multiple enterprise end-user production environments. Uniquely architected to address the plethora of quickly evolving Data-Driven applications, CSS is proven across Database, Big Data, Data Warehousing and other use cases to dramatically improve IT infrastructure efficiency. By extending low-latency PCIe 3D-NAND SSDs to address both storage I/O and algorithmically intense computational bottlenecks, CSS uniquely minimizes application run-time, optimized flash capacity utilization, and reduced total cost of ownership for advanced data center compute and storage infrastructure.

“From high transaction throughput and e-commerce payment environments to travel sites that demand real time response to queries, our customers are getting more useful work from their volume flash storage deployments with CSS,” said Hao Zhong, Co-Founder and CEO of ScaleFlux. “Leveraging programmable hardware, we are able to address new customer demands on a multi-month cycle versus a multi-year ASIC development cycle for new intelligent storage features, which puts us in an advantageous position in this fast-moving market.”

CSS easily integrates into all standard Linux / x86 server flash storage environments and has been validated to improve the efficiency of popular applications including Aerospike, PostgreSQL, MySQL, DeepgreenDB, RocksDB, HBase, Hadoop, and Spark (see http://www.scaleflux.com/applicationvalue.html for benchmark details). The list of applications continues to expand as ScaleFlux innovates from both a compute engine and Storage I/O acceleration perspective, and will extend into content delivery, search, HPC, AI and machine learning environments.

“ScaleFlux CSS has proven to deliver extremely high transactions with consistently low latency – something that is becoming more of a challenge as 3D NAND evolves to larger and larger die densities,” said Brian Bulkowski, Co-Founder & CTO for Aerospike. “We are excited to collaborate with ScaleFlux on future acceleration functions that can extend Aerospike’s value proposition as the leading, large scale Hybrid Memory Database for real-time applications.”

ScaleFlux CSS is available today in both PCIe Card and U.2 form factors with up to 6.4 Terabytes of state-of-the-art 3D NAND. Through a turnkey, easily-to-install software package, both low-latency storage I/O and compute hardware acceleration are easily enabled without development effort or application re-compilation.

“Computational Storage is an innovative solution to complex, application-level bottlenecks,” said Jim Handy, General Director of Objective Analysis. “ScaleFlux has found a way to combine this concept with the dramatic increase in flash deployment in virtually every data center.”

Come Visit ScaleFlux at the Gartner IT Data Center Conference -- Venetian Hotel, Las Vegas, NV, December 4-7, Solution Showcase Hall C Level 2, Inspur Booth #351

Industry experts and end users will come together to discuss and evaluate the latest data center IT technologies at the Gartner IT Data Center Conference. ScaleFlux will be running a live demonstration of an Aerospike NoSQL Database accelerated by low-latency CSS on an Inspur 1U, volume server. Inspur, a leading data center and cloud computing total solutions provider and worldwide top 4 largest server manufacturer, has fully integrated, tested, and validated CSS on its platform and has declared ScaleFlux a certified partner.

ScaleFlux Adds Xilinx As A Distinguished Strategic Investor ScaleFlux Adds Xilinx As A Distinguished Strategic Investor
Innovative Computational Storage Shipping Now in Inspur Servers Innovative Computational Storage Shipping Now in Inspur Servers
November 27th, 2017 | 2017年11月27日 |
via Business Wire 通过 Business Wire

SAN JOSE, California -– ScaleFlux, Inc., the pioneer in the deployment of computational storage at scale, announced today that it has added Xilinx to the company’s distinguished list of investors. ScaleFlux leverages the architectural advantages of Xilinx devices to deliver on the promise of moving computationally advanced algorithms to data to optimize infrastructure scaling and efficiency for hyperscale and enterprise data centers.

ScaleFlux ships the industry’s first production qualified Computational Storage Subsystem (CSS) products for customer deployment. Built upon the unique feature set of low-power Xilinx FPGAs combined with Terabytes of 3D NAND flash and state-of-the-art flash controller technology, CSS provides a galvanic solution for scaling compute intensive application infrastructure. CSS is accessed through turnkey Linux plug-in software packages and easily installs into servers through standard PCIe HHHL AIC and 2.5” U.2 drive level form factors.

“Empowered by leading edge UltraScale and UltraScale+ FPGA technology from Xilinx, ScaleFlux is delivering innovative SSD solutions with integrated offloads for computationally intensive tasks across a broad set of applications,” said Manish Muthal, Vice President, Data Center Business at Xilinx. “We are delighted to be a strategic investor in ScaleFlux, and look forward to enabling power efficient computational storage to Data Centers worldwide.”

Inspur, a leading data center and cloud computing total solutions provider and worldwide top 4 largest server manufacturer, has qualified the Xilinx-based ScaleFlux CSS and features the products in its server architecture optimized for data-driven applications (see here for details). This offering is a strong proof-point of the wide applicability and market potential for ScaleFlux CSS.

“Hyperscale and enterprise customers look to Inspur to provide them with the most technologically innovative platforms that scale their businesses cost-efficiently and reliably,” said Dolly Wu, VP and GM of Inspur USA Cloud Datacenter Division. “With the integration of ScaleFlux CSS into our portfolio, Inspur provides our worldwide customer base with the most efficient, data-driven server and storage platforms to deploy Aerospike, Hadoop, HBase, KV- Store databases, MySQL, Spark, and many other applications at scale.”

Inspur has fully integrated, tested, and validated CSS on its platform and has declared ScaleFlux a certified partner. ScaleFlux will demo CSS in Inspur’s booth at the Gartner IT Infrastructure, Operations, Management & Data Center Conference on December 4-7, 2017.

CSS provides hyperscale and volume 3D NAND consumers with a complete solution platform integrating value-add Computational Storage features that can be easily modified or tuned over time.

“Xilinx provides the ideal building blocks for rapid and flexible deployment of Computational Storage,” said Hao Zhong, CEO and Co-Founder of ScaleFlux. “We are excited to add Xilinx as a valued partner and investor with a shared vision on the future of volume deployment of computational storage.”

SAN JOSE, California -– ScaleFlux, Inc., the pioneer in the deployment of computational storage at scale, announced today that it has added Xilinx to the company’s distinguished list of investors. ScaleFlux leverages the architectural advantages of Xilinx devices to deliver on the promise of moving computationally advanced algorithms to data to optimize infrastructure scaling and efficiency for hyperscale and enterprise data centers.

ScaleFlux ships the industry’s first production qualified Computational Storage Subsystem (CSS) products for customer deployment. Built upon the unique feature set of low-power Xilinx FPGAs combined with Terabytes of 3D NAND flash and state-of-the-art flash controller technology, CSS provides a galvanic solution for scaling compute intensive application infrastructure. CSS is accessed through turnkey Linux plug-in software packages and easily installs into servers through standard PCIe HHHL AIC and 2.5” U.2 drive level form factors.

“Empowered by leading edge UltraScale and UltraScale+ FPGA technology from Xilinx, ScaleFlux is delivering innovative SSD solutions with integrated offloads for computationally intensive tasks across a broad set of applications,” said Manish Muthal, Vice President, Data Center Business at Xilinx. “We are delighted to be a strategic investor in ScaleFlux, and look forward to enabling power efficient computational storage to Data Centers worldwide.”

Inspur, a leading data center and cloud computing total solutions provider and worldwide top 4 largest server manufacturer, has qualified the Xilinx-based ScaleFlux CSS and features the products in its server architecture optimized for data-driven applications (see here for details). This offering is a strong proof-point of the wide applicability and market potential for ScaleFlux CSS.

“Hyperscale and enterprise customers look to Inspur to provide them with the most technologically innovative platforms that scale their businesses cost-efficiently and reliably,” said Dolly Wu, VP and GM of Inspur USA Cloud Datacenter Division. “With the integration of ScaleFlux CSS into our portfolio, Inspur provides our worldwide customer base with the most efficient, data-driven server and storage platforms to deploy Aerospike, Hadoop, HBase, KV- Store databases, MySQL, Spark, and many other applications at scale.”

Inspur has fully integrated, tested, and validated CSS on its platform and has declared ScaleFlux a certified partner. ScaleFlux will demo CSS in Inspur’s booth at the Gartner IT Infrastructure, Operations, Management & Data Center Conference on December 4-7, 2017.

CSS provides hyperscale and volume 3D NAND consumers with a complete solution platform integrating value-add Computational Storage features that can be easily modified or tuned over time.

“Xilinx provides the ideal building blocks for rapid and flexible deployment of Computational Storage,” said Hao Zhong, CEO and Co-Founder of ScaleFlux. “We are excited to add Xilinx as a valued partner and investor with a shared vision on the future of volume deployment of computational storage.”

ScaleFlux VP Thad Omura to Present at Storage Visions 2017 ScaleFlux VP Thad Omura 在 Storage Visions 2017 上发表演讲
September 25th, 2017 | 2017年9月25日 |
view Storage Visions Conference Agenda Storage Visions 网站

SAN JOSE, California -– ScaleFlux, Inc. Vice President of Business Development Thad Omura will attend Storage Visions 2017 as an invited panelist. Mr Omura will take to the stage on Monday, October 16th to discuss the industry shifting advantages delivered by the oncoming transition to computational storage. #compute2data

The official conference description for the panel is as follows:

Subject:Bringing Compute to the Data

Topic Introduction:Over the entire history of computing data has resided in storage and memory, and has been summoned to the data processing element as it is needed. Today the industry is discovering that the movement of Big Data to the processor consumes inordinate power and incurs significant time penalties. This panel, featuring leading companies who address this issue, will discuss their current efforts to move compute to the data to save power, accelerate processing speed, and even improve scalability, in order to greatly enhance the cost/performance of tomorrow’s computers.

About Storage Visions 2017

Storage Visions is a conference based in northern California featuring keynotes, sessions, and exhibits examining the latest architectures in memory and storage technologies.

圣荷西, 加州 -– ScaleFlux, Inc. 业务发展副总裁 Thad Omura 将作为受邀演讲者参加在美国加州举办的 Storage Visions 2017 会议。Mr. Omura 将于10月16日(星期一)介绍带计算的优势。#compute2data

演讲内容如下:

主题:Bringing Compute to the Data

主题介绍:Over the entire history of computing data has resided in storage and memory, and has been summoned to the data processing element as it is needed. Today the industry is discovering that the movement of Big Data to the processor consumes inordinate power and incurs significant time penalties. This panel, featuring leading companies who address this issue, will discuss their current efforts to move compute to the data to save power, accelerate processing speed, and even improve scalability, in order to greatly enhance the cost/performance of tomorrow’s computers.

关于 Storage Visions 2017

Storage Visions 是个关于研究内存和存储技术的最新架构的加利福尼亚北部会议。

Transwarp Selects ScaleFlux to Accelerate Real-Time, Big Data Platform 星环科技超融合大数据一体机选择 ScaleFlux
Computational Storage Optimizes Performance and Infrastructure Utilization of Data-Driven Integrated Solution 带计算的存储优化星环大数据方案的性能和功效
August 22nd, 2017 | 2017年8月22日 |
view Transwarp Chinese Release 星环科技中文信息

SHANGHAI, China -– Transwarp Technology, a specialized solution provider of optimized database platforms in the Big Data Era, has announced the integration of Computational Storage Subsystem (CSS) products into the Transwarp Data Hub (TDH) and Data Appliance (TxData) solutions. CSS is a ground-breaking subsystem solution provided by ScaleFlux, Inc., the pioneer in deploying Computational Storage solutions at scale.

TDH is a one-stop big data platform for data integration, storage, management, computing, machine learning, deep learning and advanced analytics. TxData is an integrated server appliance that is specially designed and optimized for TDH deployments by leveraging acceleration of SSD flash storage, FPGA technology, system optimization to achieve maximum performance of big data platform. TxData uses a stackable, modular, high-density server architecture in 2U rack space to accommodate more server nodes, and easily scaled from a 2U, 112-core, 48TB HDD/4.8TB SSD storage system to a full 32U rack of 1792 cores 1536TB HDD/76.8TB SSD storage solution.

ScaleFlux provides industry’s first, production-qualified Computational Storage Subsystem (CSS) solutions that uniquely integrate algorithmically complex compute engines with Terabytes of 3D NAND flash storage. CSS simply integrates into standard server and storage infrastructure through an easy-to-install Linux plug-in software package combined with standard PCIe HHHL Add-In Card and 2.5” U.2 drive level form factors.

“As a one-stop, easy-to-use, big data platform, TDH and the TxData combine for unprecedented real-time data analytics, streaming, and unstructured data processing performance,” said Yuanhao Sun, Founder of Transwarp. “ScaleFlux CSS easily integrates into our platform and provides 3D NAND Flash storage with unique, computational acceleration that further extends our performance value proposition vs. other big data and database offerings.”

Armed with state-of-the-art in-memory computation, efficient indexing, execution optimization, and high fault tolerance, TDH handles processing of data sets up to 10PB as a single instance. Transwarp’s value has been proven in the market by having deployed the most instances of Hadoop in the entire China market.

“Our mission is to further proliferate the value and deployment of Computational Storage in data-driven applications that face bottlenecks from both a compute efficiency and storage I/O perspective,” said Hao Zhong, CEO and co-founder for ScaleFlux, Inc. “It is an honor to work with Transwarp and their team of experts in real-time Big Data and Database applications to enable easy-to-use platforms that accelerate the time-to-insight gained from the real-time analysis of massive, multi-petabyte workloads.”

ScaleFlux CSS features field-programmable gate arrays (FPGAs) and enables acceleration for compression, erasure coding, and KV store, in addition to encryption, authentication, and common compute intensive algorithms. Future application level acceleration algorithms such as database queries, machine learning, AI, media transcoding, search, and many more, can be dropped into the same hardware infrastructure. By taking a programmable hardware approach, ScaleFlux has set CSS apart from commodity SSDs through the addition of proven, application level, value-added features while facilitating mass deployment of state-of-the-art 3D NAND Flash media.

About Transwarp Technology (www.transwarp.io)

Transwarp Technology is a leading big data software provider worldwide, focusing on enterprise database R&D and services. In 2016, Transwarp was recognized by Gartner as the most visionary company worldwide in data warehouse and data management. Transwarp is headquartered in Shanghai and has branches in more than 10 provinces and cities like Beijing, Guangzhou, Nanjing, and Jinan. Now Transwarp has finished series C round funding, led by Tencent.

2017年8月21日中国上海 -- 专业的大数据公司星环科技宣布在一站式大数据综合平台 Transwarp Data Hub(TDH)和超融合大数据一体机(TxData)方案中集成计算存储子系统(CSS)产品,ScaleFlux 是首家规模部署带计算存储解决方案的公司, CSS产品是ScaleFlux 提供的一个开创性的计算存储融合子系统解决方案。

星环科技的 TDH 是高性能自动化部署的大数据平台,一站式完成数据集成、存储、治理、计算、统计、挖掘、深度分析等工作。数据库提供。TxData是超融合大数据一体机,为TDH的部署做了特别的设计和优化,通过SSD闪存加速、FPGA加速、系统优化等方法,实现大数据处理平台性能最大化。TxData采用可堆叠、模块化的高密度服务器架构,在2U的机柜空间承载更多的服务器节点,可以轻松从2U,112核,48TB 硬盘/ 4.8TB SSD扩展到 32U, 1792核,1536TB硬盘/ 76.8TB SSD 的整机柜方案。

ScaleFlux提供行业首家产品化的计算存储子系统(CSS)的解决方案,CSS将重型计算引擎直接与3D NAND闪存存储结合,通过标准PCIe HHHL插件卡和2.5寸 U.2形态集成到服务器与存储基础架构中。

“作为一个一站式的,易于使用的大数据平台,TDH和TxData 的组合 可以提供远超传统方案的实时数据分析和非结构化数据处理的性能,”星环科技的首席技术官和创始人孙元浩说,“ScaleFlux 的CSS产品可以方便地集成到我们的平台, 提供3D NAND 闪存存储与独特的计算加速能力,进一步提高我们和其他大数据和数据库产品的的竞争力。”

结合最先进的内存计算,高效的索引,优化运行和高容错的特性,TDH可处理的高达10PB的单实例场景。星环科技的价值已经被市场认可, 在整个国内市场的部署Hadoop实例数稳居第一。

“我们的任务是进一步推进解决大数据应用和部署所面临的计算效率和存储I / O瓶颈,”ScaleFlux的首席执行官和联合创始人钟浩说,“ 我们很荣幸能和星环的专家团队在实时大数据和数据库应用上部署易于使用的平台,提高海量实时分析的效能。”

ScaleFlux CSS采用赛灵思现场可编程门阵列(FPGA),除了加密,认证和常用的计算密集型算法之外,还能够进行压缩,纠删,KV存储加速。此外,未来的平台和应用程序级加速算法,如数据库查询,机器学习,人工智能,媒体转码,搜索等等,都可以在同一基础设施中实现。拥有可编程硬件的计算能力, ScaleFlux 将CSS 产品和其他SSD 产品区别开来, 通过应用级别的优化, 增值的产品特性来进一步增值3D NAND 闪存大规模部署的价值。

关于星环科技 (www.transwarp.io)

星环科技是全球领先的大数据基础软件平台供应商,专注于企业级大数据核心平台数据库研发和服务。2016年被Gartner评为全球最具有前瞻性的数据仓库及数据管理解决方案厂商。公司总部设在上海,在北京、广州、南京、郑州等地设有分支机构。目前已完成C轮融资,腾讯领投。

ScaleFlux CEO Hao Zhong to Present at Open-Source Database Forum in Beijing ScaleFlux CEO 钟浩在北京开源数据库论坛上发表演讲
August 11th, 2017 | 2017年8月11日 |
visit ODF Website ODF 网站

SAN JOSE, California -– ScaleFlux, Inc. CEO & Co-Founder Hao Zhong will attend Open-Source Database Forum 2017 in Beijing, China as an invited speaker. Dr. Zhong will present on Friday, August 25th on the advantages offered by computational storage and the CSS product ecosystem.

The official conference description for Hao's talk is as follows:

Subject:Computational Storage System (CSS)'s speed and efficiency gains in SQl/NoSQL databases

Topic Introduction:Modern enterprise and cloud computing data centers face significant storage and computational load as a result of unprecendented growth in data sources. The three major pillars of data center infrastruture: network, storage, and computing, have evolved dramatically in the past decade. Gains in CPU frequency and multi-core development have been gradually flat, leading to data calculation and processing becoming more and more of a bottleneck. The integration of the high performance compute engines and high density flash memory of CSS will greatly enhance the speed and efficiency of heavy load calculation.

Speaker Introduction:Hao Zhong is a veteran leader of the storage industry, driving innovation with his wealth of R&D experience in such fields as hard disk r/w chips, SSD controllers, and PCIe SSD subsystems. An Electrical Engineering PhD from Rensselaer Institute of Technology, Dr. Zhong has previously served as senior technical director of Fusion-IO, technical director at SandForce, and system architect at LSI. In 2014, Dr. Zhong led a team to create ScaleFlux, Inc., a company focused on enhancing the performance of data center storage and computing, especially for database and large data applications.

About ODF 2017

The Open-Source Database Forum invites China's leading experts in large enterprise technology to share their experience and latest developments in the open source database ecosphere. ODF 2017 will be held in Beijing, China, on August 24th and 25th.

圣荷西, 加州 -– ScaleFlux, Inc. 创始人 & CEO 钟浩将作为受邀演讲者参加在中国北京举办的 Open-Source Database Forum 2017。钟博士将于8月25日(星期五)介绍CSS产品的优势。

演讲内容如下:

主题:可计算存储子系统(CSS)对SQL和NoSQL数据库的加速和效率提升

主题介绍:当今企业和云计算数据中心正面临着迅猛数据量增长和随之带来数据存储和计算负荷的重大挑战。 在过去的10年里,数据中心基础设施的三大部分:网络,存储,和计算发生了巨大变化。 在网络和存储性能实现了暴增的同时,计算方面的CPU 主频一直停滞不前,而多核的发展也渐近平缓。这使得数据计算和处理逐渐成为瓶颈,从而限制的服务器和存储的进一步提升。 融合了高性能计算引擎和高密度闪存的CSS , 大幅提升重负荷计算的效率。

嘉宾介绍:美国Rensselaer理工学院电子工程系博士;曾任美国Fusion-IO公司高级技术总监,SandForce公司技术总监和LSI公司系统架构师。拥有丰富的数据存储产品架构,创新和研发的实战经验,多次引领存储行业在硬盘读写芯片,SSD控制器及PCIe SSD子系统的技术革新 。2014年钟浩博士率领团队创建ScaleFlux公司,致力于提升数据中心存储和计算的性能与效率的提升,尤其是对数据库和大数据的应用进行了大幅度的优化。

关于 ODF 2017

开源数据库论坛拟邀请中国顶尖互联网公司和大型企业技术专家分享其在开源数据库方向上的经验以及最新进展。2017年ODF将于8月24日和25日在北京举行。

ScaleFlux Showcase at Flash Memory Summit 2017 锐开在 FMS 2017
August 7th, 2017 | 2017年8月7日 |
via PRNewswire 来自 PRNewswire

SANTA CLARA, California -- ScaleFlux will be in attendance at Flash Memory Summit 2017 to showcase its product at Xilinx, booth #126, August 8-10, 2017 at the Santa Clara Convention Center, Santa Clara, California.

ScaleFlux Computational Storage Subsystem (CSS) uniquely solves both compute and storage I/O bottlenecks. CSS compression accelerates throughput by an order-of-magnitude without the costly CPU overhead incurred by software solutions, resulting in optimal storage capacity utilization without performance compromises.

UCloud Selects ScaleFlux for Data-Driven Cloud Infrastructure UCloud选择ScaleFlux技术构建数据驱动的云基础架构
August 7th, 2017 | 2017年8月7日 |
via Business Wire | view UCloud Chinese Release 通过 Business Wire | 看UCloud中文信息

SAN JOSE, California -- ScaleFlux, Inc., the pioneer in deploying Computational Storage at scale, announced that UCloud, China's leading independent public cloud provider, has selected ScaleFlux to power its advanced UPHost (UCloud Physical Host) infrastructure for Big Data, Database, Data Warehousing, and a fast-growing list of data-driven applications.

ScaleFlux provides industry’s first, production-qualified Computational Storage Subsystem (CSS) solutions that uniquely integrate algorithmically complex compute engines with Terabytes of 3D NAND flash storage. CSS simply integrates into standard server and storage infrastructure through an easy-to-install Linux plug-in software package combined with standard PCIe HHHL Add-In Card and 2.5” U.2 drive level form factors.

“UPHost offers outstanding computing performance to meet the core application scenarios for high performance and stability, and is built to run applications at any scale for our customers,” said Xianfeng Mo, CTO and Founder of UCloud. “With the integration of ScaleFlux CSS into UPHost, we are further extending UCloud’s value proposition by dramatically reducing the time to complete Big Data and Database jobs, while also optimizing flash storage capacity utilization.”

Large pools of Computational Storage will eliminate the scaling roadblocks faced by data-driven infrastructure where:

  1. traditional CPUs are highly inefficient when tasked to process modern, intense workloads
  2. the latency and power to move massive datasets from storage to the CPU is too expensive
  3. compute-intensive workloads can be sped up by orders-of-magnitude when parallelized across flash storage subsystems

These scaling challenges are commonly faced with applications in the field of Big Data, Database, Data Warehousing, Content Delivery, Search, Machine Learning, AI, VR, Genomics, Security, High-Performance Computing, and many others.

“By taking an evolutionary and simple approach to integrating CSS into application frameworks, we are delighted to provide our revolutionary Computational Storage solutions for UCloud’s UPHost infrastructure that will open new revenue generating use-cases for data-driven applications at scale with fast time-to-market,” said Hao Zhong, CEO and co-founder for ScaleFlux, Inc. “One of the primary values of CSS is the ability to reprogram new hardware acceleration engines in-system, which can evolve the infrastructure in lock step with new customer demands.”

ScaleFlux CSS features field-programmable gate arrays (FPGAs) and enables acceleration for compression, erasure coding, and KV store, in addition to encryption, authentication, and common compute intensive algorithms. Future application level acceleration algorithms such as database queries, machine learning, AI, media transcoding, search, and many more, can be dropped into the same hardware infrastructure. By taking a programmable hardware approach, ScaleFlux has set CSS apart from commodity SSDs through the addition of proven, application level, value-added features while facilitating mass deployment of state-of-the-art 3D NAND Flash media.

ScaleFlux CSS is available for customer deployment today.

About UCloud (www.ucloud.cn)

UCloud (Shanghai UCloud Cloud Computing Co., Ltd.) is a leading IaaS service provider in China. UCloud's own R&D solutions range from computing and storage to network resources and other fundamental IT infrastructure that enterprises need. UCloud has provided services to 50,000 enterprise customers worldwide.

日前,UCloud宣布将与ScaleFlux公司合作,将ScaleFlux提供的计算存储子系统(CSS)产品整合到UCloud基础设施中。CSS将重型计算引擎直接与3D NAND闪存存储结合,通过标准PCIe HHHL插件卡和2.5 U.2形态集成到服务器与存储基础架构中。该产品后续将应用于UCloud服务器产品,在混合云、大数据、内存存储、数据库等业务上发挥作用。

众所周知,SSD的吞吐量是机械硬盘的10-20 倍,随机读写性能更是可高达机械硬盘的数千倍,这对数据中心和云基础设施产生了深远影响。随着闪存成为工作负载应用程序的主流存储介质,加上引入性价比更高的3D NAND介质,存储性能增大只会加剧内存I/O和容量受限的问题。而可计算存储 (Computational Storage Subsystem)的部署标志着IT基础架构解决方案中的根本变化,该技术可以解决计算和存储I/O双密集应用的巨浪冲击。

UCloud CTO莫显峰表示: "Uloud提供高效的云基础设施,为客户各类应用程序提供服务。将ScaleFlux CSS整合到UCloud基础设施平台中,不仅有利于提升闪存存储空间利用率,而且在特定业务场景下可以利用CSS本身的计算性能节省CPU资源。”

据介绍,大型计算存储池可以消除数据驱动型应用程序面临的扩展障碍,应用了CSS技术,CPU处理现代复杂工作负载的性能限制将被突破;海量数据使得从存储到CPU移动数据产生巨大延时和功耗开销将减小;大数据、数据库、内容交付、搜索、机器学习、人工智能、虚拟现实、基因组学以及安全高性能计算等计算密集型工作负载,可以实现多个数量级的加速。

“我们很高兴为UCloud领先的云基础架构提供革命性的计算存储CSS解决方案,通过采用革命性但简单的方法将CSS集成到现有的IT基础架构和应用程序框架中,为数据驱动的应用程序开发新的用户场景和用例。”ScaleFlux公司首席执行官兼联合创始人钟浩表示,“CSS的主要价值就是在系统级别上将硬件加速引擎的重新编程能力引入到平台中,与客户不断发展的新需求保持一致。”

ScaleFlux CSS采用赛灵思现场可编程门阵列(FPGA),除了加密、认证和常用的计算密集型算法之外,还能够进行压缩、纠删、KV存储加速。此外,未来的平台和应用程序级加速算法,如数据库查询、机器学习、人工智能、媒体转码、搜索等,都可以在同一基础设施中实现。