Intel and Lightbits Labs team up to improve NVMe/TCP performance

Read Time:3 Minute, 15 Second

The collaboration between Intel and Lightbits Labs aims to improve the start-up’s NVMe/TCP storage solution by moving away from Fiber Channel and optimizing the use of hardware resources.

Intel makes NVMe-oF/TCP shine with new partnership

Intel has partnered with Lightbits Labs (investing in the startup alongsidee Dell and Cisco), to improve the performance of storage systems in data centers. The two companies plan to develop disaggregated solutions with the aim of reducing the total cost of ownership (TCO) of storage systems, in particular by limiting unnecessary capacity allocations. By refining the allocation of storage resources, Lightbits intends to allocate just what is needed by applications and VMs. The end result is – in theory at least – less power consumption since unused space is no longer allocated. It is always difficult – unless you have dedicated analytical tools – to optimize the provisional allocation of storage resources. However, the cloud can contribute to refining this management of available or reserved space in hybrid mode.

Based in San Jose, CA, and founded in 2016 in Kfar Saba, in Israel, the start-up considers itself the “creator” of NVMe-over-TCP because it contributed code and worked alongside Facebook, Intel, Cisco, Dell EMC and Micron to develop the NVMe-over-TCP. Standard TCP, which was ratified by the NVM Express consortium in late 2018 as the link transport layer. Lightbits also aims to improve TCO by eliminating the need for specialized hardware. With LightOS NVMe over Fabrics TCP (NVMe-oF/TCP), the start-up has in fact developed a software-defined block mode storage platform that groups NVMe flash storage space distributed over several nodes on Ethernet using the TCP/ IP rather than a more expensive dedicated network such as Fiber Channel.

Faster NVMe SSDs

Intel, which markets NVMe SSDs with NAND flash or Optane components, is probably not the first name that comes to mind in the small world of storage, but it works with the main suppliers. LightOS, while fully optimized for Intel hardware, will provide customers with improved storage efficiency, reduce underutilization, and maintain compatibility with existing infrastructure without compromising performance.

“Data centers are undergoing a transformation, with resource disaggregation and composability essential to meet the efficiency requirements needed to address the data explosion,” said Rémi EL-Ouazzane, vice president and Director of Strategy and Business Development for the Data Platform Group at Intel, in a press release. “Our differentiated hardware capabilities combined with Lightbits’ innovative NVMe over Fabrics software provide our joint customers with a cost-effective and exceptional solution to address this strategic inflection point.”

Intel gear to get started

The LightOS storage solution was tested with Intel’s Ethernet 800 Network Adapters with Application Device Queues (ADQ), designed to achieve faster, more predictable Ethernet. ADQ enables the NVMe-oF/TCP interface to achieve distributed storage performance in the same range as RDMA-based protocols, while NVMe-oF/TCP enables wide adoption due to its ease of deployment and its scalability. Intel claims that LightOS with ADQ on the Ethernet 800 card shows up to 30% improvement in response time predictability measured by P99.99 tail latency, up to 50% reduction in average latency, and up to 70% increase in throughput measured in IOPS when using ADQ vs without ADQ. In addition to the Intel Ethernet adapter, LightOS will also be tailored to work with Intel’s Optane persistent storage and 3D NAND (QLC) SSDs, as well as its Xeon controllers.

Lightbits Labs isn’t the first NVME-oF storage startup Intel has invested in. This was already the case for the promising young shoot E8 Storage, which we met several times as part of the IT Press Tour, and which ended up in the hands of AWS. , but did not disclose financial details of its investment. Protocols such as NVMe and NVMe-oF enable ultra-fast, low-latency flash storage, which means enterprises can run distributed applications, including Kubernetes, and access data in near real-time.

Happy
Happy
0 %
Sad
Sad
0 %
Excited
Excited
0 %
Sleepy
Sleepy
0 %
Angry
Angry
0 %
Surprise
Surprise
0 %
Samsung Adapts To Hybrid Working Previous post Samsung adapts to hybrid working
With Purview, Microsoft Unifies Data Management Based On Compliance Next post With Purview, Microsoft unifies data management based on compliance