Jump to content

VAST Data: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
Fxober (talk | contribs)
m Technology: small changes to how lightspeed was defined, more descriptive to the 3 core elements
Fxober (talk | contribs)
Line 39: Line 39:
VAST Data, with regard to universal storage, uses Flash-QLC and 100 percent persistent global namespace.<ref>{{cite web |title=Vast Data |url=https://www.sdxcentral.com/articles/news/vast-data-raises-80m-for-flash-performance-at-hard-disk-prices/2019/02/ |website= |date=February 27, 2019}}</ref> It supports [[artificial intelligence]] applications.<ref>{{cite web |last1=Evans |first1=Chris |title=Leading storage for AI tools address workload capacity, performance |url=https://searchstorage.techtarget.com/feature/Leading-storage-for-AI-tools-address-workload-capacity-performance |website=[[TechTarget|SearchStorage]] |date=November 13, 2019}}</ref>
VAST Data, with regard to universal storage, uses Flash-QLC and 100 percent persistent global namespace.<ref>{{cite web |title=Vast Data |url=https://www.sdxcentral.com/articles/news/vast-data-raises-80m-for-flash-performance-at-hard-disk-prices/2019/02/ |website= |date=February 27, 2019}}</ref> It supports [[artificial intelligence]] applications.<ref>{{cite web |last1=Evans |first1=Chris |title=Leading storage for AI tools address workload capacity, performance |url=https://searchstorage.techtarget.com/feature/Leading-storage-for-AI-tools-address-workload-capacity-performance |website=[[TechTarget|SearchStorage]] |date=November 13, 2019}}</ref>


Its 2020 storage architecture release, known as LightSpeed, combines three core elements to deliver parallel and higher scalable performance for AI and HPC. These elements are:
Its Year 2020 storage architecture release, known as LightSpeed, combines three core elements to deliver parallel and higher scalable performance for AI and HPC. These elements are:
1. RDMA for NFS version 3.
1. RDMA for NFS version 3.
2. nconnect to create multiple network paths to a single NFS mount.
2. nconnect to create multiple network paths to a single NFS mount.

Revision as of 00:42, 30 December 2020

VAST Data
Company typePrivately held company
IndustryData storage
FoundedJanuary 2016; 8 years ago (2016-01)
Headquarters,
U.S.
Key people
Renen Hallak
Shachar Fienblit
Jeff Denworth
Mike Wing
Websitevastdata.com

VAST Data is a technology company that focuses on data storage, specifically flash memory.[1][2] Founded in 2016, VAST has offices in the United States and Israel, with its headquarters in New York.[3] VAST Data was founded with the aim of replacing multiple storage tiers with one solid state platform.[4][5]

VAST Data is being used by the National Institutes of Health and Harvard University to combat the COVID-19 pandemic, as well as by Ginkgo Bioworks for genomic studies.[6][7]

History

VAST was founded in 2016 by Renen Hallak, a former engineer of R&D at XtremIO, Shachar Fienblit, formerly at Kaminario and Jeff Denworth, formerly at CTERA Networks.[7][3] Mike Wing of Dell is a part of the leadership team at VAST Data.[3]

VAST Data's launch was supported by Dell and Goldman Sachs.[8][9][10]

As of April 2020, the company has 145 employees.[7] During the COVID-19 pandemic, most employees are working from home.[11][12]

Technology

At the front end of its storage, VAST Data uses Intel's Optane (3D X-Point) NVMe SSDs.[4][13][5] 3D XPoint non-volatile memory is integrated in VAST Data's architecture.[14]

VAST Data's technology allows for collapsing multiple storage tiers into one that has decoupled compute nodes, which are accessed using NVMe-oF.[1] After data reduction occurs, around 2PB of space is available.[1] The single VAST Data tier uses wide data stripes, with the purpose of global erasure coding.[1] NVMe-linked Databoxes contain flash drives for data and Optane XPoint for metadata.[1] VAST storage enclosures are connected to servers through NVMe-oF, using either 100Gb/s Ethernet or Infiniband.[15]

VAST Data, with regard to universal storage, uses Flash-QLC and 100 percent persistent global namespace.[16] It supports artificial intelligence applications.[17]

Its Year 2020 storage architecture release, known as LightSpeed, combines three core elements to deliver parallel and higher scalable performance for AI and HPC. These elements are: 1. RDMA for NFS version 3. 2. nconnect to create multiple network paths to a single NFS mount. 3. NVidia GPUDirect available for NVIDIA GPU[18][19] [20]

References

  1. ^ a b c d e Mellor, Chris (February 26, 2019). "VAST Data: The first thing we do, let's kill all the hard drives". Blocks and Files.
  2. ^ Miller, Ron (April 16, 2020). "TechCrunch".
  3. ^ a b c Mellor, Chris (February 26, 2019). "VAST Data's business sitrep". Blocks and Files.
  4. ^ a b Coughlin, Tom (March 1, 2019). "VAST Storage Dreams". Forbes.
  5. ^ a b Adshead, Antony. "QLC + 3D Xpoint = Vast Data's universal storage". Computer Weekly.
  6. ^ Williams, Joe (April 17, 2020). "Business Insider".
  7. ^ a b c Wilson, Alexandra (April 16, 2020). "VAST Data Gets $1.2 Billion Valuation, As Researchers Race To Combat COVID-19". Forbes.
  8. ^ Sawaya, Sydney (April 16, 2020). "Vast Data Grows War Chest". SDxCentral.
  9. ^ Deutscher, Maria (April 16, 2020). "SiliconANGLE".
  10. ^ Deutscher, Maria (September 12, 2019). "With upgraded storage platform, Dell-backed Vast Data targets AI workloads". SiliconANGLE.
  11. ^ Kovar, Joseph F. (April 16, 2020). "Renen Hallak Talks Future". CRN.
  12. ^ Costa, Cameron (May 5, 2020). "CNBC".
  13. ^ Marko, Kurt (September 25, 2019). "NVMe-oF products are changing the face of data storage". TechTarget.
  14. ^ "Vast Data Adds Zero Data Loss Guarantee for Storage". Channel Futures. June 7, 2019.
  15. ^ Harris, Robin (July 27, 2020). "Vast Data: A smart new architecture for cloud scale storage". ZDNet.
  16. ^ "Vast Data". February 27, 2019.
  17. ^ Evans, Chris (November 13, 2019). "Leading storage for AI tools address workload capacity, performance". SearchStorage.
  18. ^ Kovar, Joseph F. (September 15, 2020). "Vast Data Aims At Artificial Intelligence With LightSpeed Architecture". CRN.
  19. ^ Coughlin, Tom (September 23, 2020). "Cloud At The Edge, GPU Storage And LTO Gen 9". Forbes.
  20. ^ Denworth, Jeff (December 14, 2020). "LightSpeed Universal Storage for A.I. Webinar — VAST Data". YouTube.