Kalray Unveils its K200-LP Latest Acceleration Card for Data Centers
Grenoble – France, June 16, 2021 – Kalray, a leading provider in the new generation of processors specialized in Intelligent Data Processing from Cloud to Edge, today unveils its new K200‐LP™ acceleration card. K200‐LP™ is a new member of Kalray’s fully programmable multi‐purpose acceleration card family, based on its unique MPPA® processor, and offering an unprecedented level of performance and targeting the booming NVMe based storage market as its first implementation.
The explosion of data, driven by demanding services and applications using AI and data analytics, is putting a lot of pressure on Cloud and Edge Data Center’s networking and storage to feed the demanding compute intensive resources. The primary needs are to reduce latency, enhance bandwidth and IOPs(1) performance as well as offloading host CPUs, at best cost.
Data centers are looking for solutions that enable high performance, low power, cost effectiveness and openness to scale and match demand on data availability. The K200‐LP™ acceleration card is Kalray’s solution to address these exacting demands for their data center customers.
Eric Baissus, President and CEO of Kalray, said: “We are proud to unveil our new K200‐LP™ acceleration card, based on our MPPA® processor. This family of Ethernet/PCIe acceleration cards are natively capable of managing multiple workloads with no bottleneck to enable smarter, more efficient, and energy‐wise applications on Cloud and Edge data centers. To address the new generation of storage array solutions for Cloud and Edge, K200‐LP™ is a game changer solution in terms of performance per Watt and per dollar.”
To improve and optimize overall IT infrastructures, the industry has moved increasingly towards disaggregated storage. Put simply, separating compute resources and storage resources.
Most advanced All‐Flash‐Array solutions today use multiple adapter cards around a x86 that runs storage services. Kalray manages to run all the critical functions of a disaggregated storage appliance on one single chip: the Kalray MPPA® processor. Advanced Data Processing Units (DPU) like MPPA® and associated acceleration cards can run, efficiently and on a cost‐optimized manner versus traditional x86, new protocols such as NVMe and NVMe‐over‐Fabric to access and unleash the full capability of SSDs wherever they are located in the Data Center as if there were locally attached.
Whereas the industry has defined a new NVMe protocol to support the latest generation of ultra‐fast storage devices, existing traditional technologies are not capable of exposing the full capability of NVMe‐based SSDs. Kalray’s K200‐LP™ smart storage card has been built from the ground‐up to support those breaking through data protocols. Based on Kalray MPPA’s processor, K200‐LP™ is also the perfect solution for storage array appliance makers and cloud service providers, to build their next generation of storage appliances in terms of performance per Watt per dollar.
Kalray’s K200‐LP™ acceleration card is a low‐profile, 2x 100Gb/s ethernet, PCIe Gen4 card that can deliver more than 2 MIOPS and 12 GB/s per card (both RoCE and TCP) with a latency as low as 30 microseconds. K200‐LP™ has been developed by Kalray and is manufactured by Taiwanese Wistron, one of the largest manufacturers of electronic cards and data center servers in the world.
Kalray’s K200‐LP™ smart storage card and associated tools are available now.
(1) IOPS is the numbers of Input/Output operations per second and is a performance measurement used to characterize computer storage devices.
Related Semiconductor IP
- AES GCM IP Core
- High Speed Ethernet Quad 10G to 100G PCS
- High Speed Ethernet Gen-2 Quad 100G PCS IP
- High Speed Ethernet 4/2/1-Lane 100G PCS
- High Speed Ethernet 2/4/8-Lane 200G/400G PCS
Related News
- Kalray and Arm to collaborate to bring data intensive processing and AI acceleration DPU solutions to the global Arm ecosystem
- Kalray and Pliops enters into exclusive negotiations to create a global leader in data accelerators for AI and storage acceleration
- Xilinx Demonstrates FPGA-Based Acceleration Technology for Next-Generation Data Centers at IBM Impact 2014
- Altera and Baidu Collaborate on FPGA-based Acceleration for Cloud Data Centers
Latest News
- HPC customer engages Sondrel for high end chip design
- PCI-SIG’s Al Yanes on PCIe 7.0, HPC, and the Future of Interconnects
- Ubitium Debuts First Universal RISC-V Processor to Enable AI at No Additional Cost, as It Raises $3.7M
- Cadence Unveils Arm-Based System Chiplet
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers