Upgrade to Pro — share decks privately, control downloads, hide ads and more …

VAST Data - IT Press Tour Jan 2021

VAST Data - IT Press Tour Jan 2021

The IT Press Tour

February 17, 2021
Tweet

More Decks by The IT Press Tour

Other Decks in Technology

Transcript

  1. V A S T D A T A 2 0

    2 1 I T P T 1
  2. V A S T D A T A 2 0

    2 1 I T P T 2 • VAST Data Company Focus • Product Refresher • Updates Since Last ITPT • Introducing: Peter Gadd • Financial Update • Customer Updates • Our Direction EMBARGO UNTIL 9AM ET Feb 17th
  3. SIMPLIFY STORAGE AND DEMOCRATIZE FAST ACCESS TO ALL DATA. P

    A S T 3 V A S T D A T A 2 0 2 1 I T P T
  4. 5 V A S T D A T A 2

    0 2 1 I T P T R I P MECHANICAL MEDIA STORAGE TIERING
  5. V A S T D A T A 2 0

    2 1 I T P T 6 BREAKING 30 YEARS OF INFRASTRUCTURE TRADEOFFS
  6. 7 V A S T D A T A O

    V E R V I E W
  7. ONE UNIVERSAL DATA PLATFORM FOR ALL YOUR APPLICATIONS. Write at

    3D XPoint speeds, Read at TB/s, Millions of IOPS ALL-NVMe PERFORMANCE Storage exceeding the capacity needs of any size organization EXABYTE-SCALE NAS & OBJECT Engineered at every level to deliver unrivaled system efficiency TIER-5 COST EFFICIENCY V A S T D A T A O V E R V I E W 8 V A S T D A T A O V E R V I E W
  8. 3D XPOINT QLC FLASH NVMe ENCLOSURE NVME FABRIC: COMMODITY ETHERNET

    OR INFINIBAND CLIENTS: NFS, NFS+RDMA+GPUDirectTM, SMB, S3, K8S CSI 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE G L O B A L N A M E S P A C E VAST CONTAINER SERVERS 1 0 V A S T D A T A O V E R V I E W
  9. S E G M E N T B Y A

    P P L I C A T I O N O R N E T W O R K 3D XPOINT QLC FLASH NVMe ENCLOSURE NVME FABRIC: COMMODITY ETHERNET OR INFINIBAND 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE G L O B A L N A M E S P A C E S S E G M E N T B Y A P P L I C A T I O N O R N E T W O R K ONE STORAGE SYSTEM TO CONSOLIDATE NOISY NEIGHBORS & ISLANDS OF STORAGE V A S T D A T A O V E R V I E W 1 1 V A S T D A T A O V E R V I E W
  10. 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH

    NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE 3D XPOINT QLC FLASH NVMe ENCLOSURE G L O B A L N A M E S P A C E STATELESS ARCHITECTURE, ELIMINATES EAST-WEST TRAFFIC N o C a c h e C o h e r e n c y C h a l l e n g e s | N o B a t t e r i e s | N o R e b u i l d s D u e T o S e r v e r F a i l u r e | N - W a y R e s i l i e n c e A FOUNDATION FOR GLOBAL ALGORITHMS G l o b a l N a m e s p a c e | Q L C F l a s h Tr a n s l a t i o n | G l o b a l E r a s u r e | D a t a R e d u c t i o n 1 0 µ s D I S T A N C E T O S T O R A G E S Y S T E M S T AT E 1 2 V A S T D A T A O V E R V I E W
  11. 2 X T B 2 X G B / s

    T B G B / s 4 X T B 4 X G B / s G L O B A L N A M E S P A C E G L O B A L N A M E S P A C E G L O B A L N A M E S P A C E T O D AY . . . 2X CPU 2X CPU . . . T O M O R R O W & B E Y O N D … 4X CPU 4X CPU . . . V A S T D A T A O V E R V I E W 1 3 N V M e F A B R I C N V M e F A B R I C N V M E F A B R I C
  12. S3 SMB v2 NFS v3 tba PROTOCOL LAYER DATA STORE:

    CAPACITY-EFFICIENT BLOCKS IN QLC A BYTE-GRANULAR, THIN-PROVISIONED, SHARDED, INFINITELY-SCALABLE DATA STORE ELEMENT STORE ATTRIBUTES: RICH, SHARED-EVERYTHING METADATA IN 3D XPOINT POSIX & S3 METADATA, MULTI-PROTOCOL MAPPINGS, LINKS, LOCK STATE, SNAP STATE, KEYS 1 4 V A S T D A T A O V E R V I E W
  13. Global QLC Flash Translation Global Erasure Codes Global Data Reduction

    Q L C F L A S H S AV E S 8 5 % S AV E U P TO A N OT H E R 6 6 % U N P R E C E D E N T E D E F F I C I E N C Y MLC/ TLC QLC COST 1 0 - Y E A R W A R R A N T Y 1 5 V A S T D A T A O V E R V I E W
  14. LIFE SCIENCE 2:1 ANIMATION 3:1 SEARCH 4:1 BACKUP APPLIANCES 3:1

    pre-compressed and pre-deduplicated pre-compressed pre-compressed pre-compressed un-compressed MARKET DATA 8:1 BACKUP 20:1 un-compressed HPC 3:1 pre-compressed 1 6 V A S T D A T A O V E R V I E W
  15. V A S T D A T A 2 0

    2 1 I T P T 1 7
  16. V A S T D A T A 2 0

    2 1 I T P T 1 8 MANUFACTURING NATIONAL SECURITY WEB HOSTING ENTERPRISE BACKUP/ARCHIVE VMWARE & CONTAINERS SEISMIC PROCESSING BROADCASTING SUPERCOMPUTING FILM ANIMATION QUANTITATIVE TRADING DEEP LEARNING / AI BRAIN RESEARCH GENOMICS / LIFE SCIENCE 2018 2019 2020 HORIZONTAL USE CASE VERTICAL USE CASE
  17. V A S T D A T A 2 0

    2 1 I T P T 1 9 A N N O U N C E D I N A P R I L : H I R I N G 1 0 0 P E O P L E I N 2 0 2 0
  18. JUNE OUR FIRST ITPT 2 0 V A S T

    D A T A 2 0 2 1 I T P T JULY NOV DEC AUG SEPT OCT JAN
  19. JUNE 2 1 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC AUG 5X Y/Y Q2 GROWTH WELCOME TOM MENDOZA FMR PRESIDENT & VICE CHAIRMAN OF NETAPP SEPT OCT JAN
  20. JUNE 2 2 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC AUG SEPT LIGTHSPEED FOR AI 2X AI PERFORMANCE - GPUDIRECT™ STORAGE READY OCT JAN
  21. JUNE 2 3 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC AUG SEPT OCT VAST CEO: RENEN HALLAK 100 MOST INTRIGUING ENTREPRENEURS JAN
  22. JUNE 2 4 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC AUG SEPT OCT COVID PANEL PETER GADD DANIEL BOUNDS DAVID NOY VP INTERNATIONAL VP MARKETING VP PRODUCT JAN
  23. V A S T D A T A 2 0

    2 1 I T P T 2 5
  24. V A S T D A T A 2 0

    2 1 I T P T 2 6
  25. JUNE 2 7 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC JAN AUG SEPT OCT FIRST $10M+ CUSTOMER 2nd $10M+ CUSTOMER
  26. JUNE 2 8 V A S T D A T

    A 2 0 2 1 I T P T JULY NOV DEC JAN AUG SEPT OCT 1ST HIGH- THROUHGPUT REFERENCE ARCHITECTURE
  27. 2 9 V A S T D A T A

    2 0 2 1 I T P T MAY JUN FEB MAR APR HELEN PROTOPAPAS RICK FRANKE TOM WHALEY VP FINANCE VP CUSTOMER SUCCESS VP SALES JAN
  28. 3 0 V A S T D A T A

    2 0 2 1 I T P T TOM WHALEY VP SALES RICK FRANKE VP CUSTOMER SUCCESS ASSAF COHEN VP FINANCE (ISRAEL) MICHAEL WING PRESIDENT RENEN HALLAK CEO & FOUNDER SHACHAR FIENBLIT VP R&D & CO-FOUNDER AVERY PHAM VP OPERATIONS JEFF DENWORTH CMO & CO-FOUNDER PETER GADD VP INTERNATIONAL DANIEL BOUNDS VP MARKETING DAVID NOY VP PRODUCT VP FINANCE (AMERICA) HELEN PROTOPAPAS
  29. V A S T D A T A 2 0

    2 1 I T P T 3 1 RADAR FOR FILE AND OBJECT STORAGE DEMOCRATIZING FLASH FOR MUCH BROADER WORKLOAD CONSOLIDATION IN ENTERPRISE ENVIRONMENTS (2020) FIRST ‘COOL VENDOR’ IN STORAGE SINCE 2018 FIRESTARTER AWARD RESEARCH MAP FOR FILE STORAGE MOST INNOVATIVE FLASH MEMORY TECHNOLOGY STORAGE TRAILBLAZERS AWARD WINNER
  30. V A S T U N I V E R

    S A L S T O R A G E : O N E P L A T F O R M F O R A L L D A T A C E N T E R D A T A S E C O N D A R Y S T O R A G E B I G D A T A & A I V E R T I C A L S O L U T I O N S Genomics Finance Content HPC E N T E R P R I S E I N F R A S T R U C T U R E V A S T D A T A 2 0 2 1 I T P T 3 2
  31. T H E S TO R AG E P L

    AT F O R M F O R T H E N E X T D E C A D E O F M AC H I N E I N T E L L I G E N C E . 3 3
  32. LIGHTSPEED IS NOT ONLY A PRODUCT. IT’S A MACHINE LEARNING

    STORAGE PHILOSOPHY. THE LIGHT-TOUCH SIMPLICITY OF NAS THE SPEEDTO SUPPORT ANY AI AMBITION 3 4
  33. 3 5 “Storage systems, such as those that support Lustre

    and burst buffers, are not designed for and often perform poorly for these read-heavy, random access workloads” “The newer NVMe SSDs and storage-class memory can ameliorate the low-bandwidth and high- latency challenges associated with neural networks, while the quad-level cell (QLC) drives offer flexibility to balance costs relative to sheer volume of data to be stored“ AI for Science Report, US Department of Energy Prepare Your Storage and Data Management Strategy for the Impact of Artificial Intelligence Workloads, Gartner
  34. Up to 200GB/s per Client Linear Scaling to Exabytes NAS

    (over RDMA) Simplicity 80% Less Cost N O T R A D E O F F S THE NEW AI STORAGE PARADIGM T O D A Y ’ S S H A D O W A I O P T I O N S Both Are Prohibitively Expensive for All-Flash 3 6 V A S T D A T A 2 0 2 1 I T P T U N I V E R S A L S TO R AG E . FA S T, S I M P L E , A F F O R D A B L E . H P C F I L E S Y S T E M S Fast & Scalable, But Complex S C A L E - O U T N A S Simple & Versatile, But Slow
  35. - 0 0 0 0 1 1 0 20 40

    60 80 100 120 140 160 180 T H R O U G H P U T T O G P U Throughput (GiB/s) CPU Cores Required V A S T N F S ( R D M A w M U L T I P AT H ) V A S T N F S w / G P U D i r e c t Ⓡ L E G A C Y N F S ( T C P S I N G L E P AT H ) 1 % 1 6 2 G i B / s 5 0 % 1 4 % 4 6 G i B / s 2 G i B / s SATURATE YOUR GPUS FREE YOUR X86 CORES NFS = SIMPLE AI 3 7 DGX-A100 BENCHMARKING 80X FASTER NFS PERFORMANCE TESTED WITH GDSIO • 1 x DGX-A100 • 5 x VAST CBOX SERVER CHASSIS & 5 x VAST LIGHTSPEED ENCLOSURES • 4MB I/O SIZE • 4GB FILE SIZE • 96 THREADS x 8GPUs
  36. + S U C C E S S S T

    O R Y MARTINOS TRAINS NEXT-GEN AI-BASED IMAGING WITH UNIVERSAL STORAGE 25% Less than hybrid HDD+Flash Consistent all-flash derisks GPU More GPUs Less Storage Costs = More Computing $s V A S T D A T A O V E R V I E W 3 8
  37. V A S T D A T A 2 0

    2 1 I T P T 3 9 UNIVERSAL STORAGE
  38. + DUG BUILDS MULTI-TENANT, EXASCALE INFRASTRUCTURE WITH VAST No Headaches

    vs. the complexity of Lustre Affordable IOPS 2:1 Data Reduction No Noisy Neighbors thanks to the VAST DASE architecture V A S T D A T A O V E R V I E W 4 0
  39. KRYSTAL PARTNERS WITH VAST TO CREATE UK’s HOMEGROWN HYPERCLOUD All-Flash

    WordPress & IaaS File Storage API-Driven for webscale operations Consolidated VAST pools provide QoS to competing workloads + V A S T D A T A O V E R V I E W 4 1
  40. + SQUAREPOINT CONSOLIDATES AND ACCELERATES BACKTESTING WITH VAST All-NVMe no

    more HDD access issues K8S Ready for containerized workloads No HPC Complexity A resilient and reliable production workhorse 4 2 V A S T D A T A O V E R V I E W
  41. + NIH ACCELERATES PIPELINE BY 25% ON TOP100 RESEARCH SUPERCOMPUTER

    90% Faster than their HDD GPFS systems Simple robust NAS appliance HDD Economics 2:1 Data Reduction on Genomes, Cryo V A S T D A T A O V E R V I E W 4 3
  42. V A S T D A T A 2 0

    2 1 I T P T 4 4 FASTEST-GROWING DATA INFRASTRUCTURE COMPANY EVER
  43. N D A C O N F I D E

    N T I A L 4 5 EXITED Q4 OF LAST YEAR AT $150M RUN RATE LAST YEAR WAS 3.5x OF PREVIOUS DOZENS OF $1M+ UNIVERSAL STORAGE CUSTOMERS VAST DATA ACHIEVED CASH FLOW POSITIVITY IN 2020 UNDER EMBARGO MOST SOLD IN 1ST TWO YEARS HYPERGROWTH + A BIG REVENUE BASE THE ADVANTAGE OF DIFFERENTIATION SOME OVER $10M
  44. V A S T D A T A 2 0

    2 1 I T P T 4 6 COMPARISON OF CUMULATIVE SALES TO DATE SOURCE: S-1 FILINGS, NEWS HEADCOUNT @ END OF Y2 130 81 237 350 186
  45. V A S T D A T A 2 0

    2 1 I T P T 4 7 HIRING IN ALL MAJOR REGIONS AND ACROSS ALL FUNCTIONS UNDER EMBARGO
  46. V A S T D A T A 2 0

    2 1 I T P T 4 8 UNDER EMBARGO
  47. V A S T D A T A 2 0

    2 1 I T P T 4 9 ANNOUNCING OUR THIRD $10M+ CUSTOMER MULTIPLE ORDERS FROM A US DEFENSE AGENCY UNDER EMBARGO
  48. • UNRIVALED PERFORMANCE, SCALE, SIMPLICITY & RESILIENCE • OUR HEAVY

    FOCUS ON QA DRIVES CUSTOMER SATISFACTION AND REPEATS • CUSTOMERS MAKING MAJOR INVESTMENTS OUT OF THE GATE • RESULTING IN: FASTEST GROWING COMPANY IN OUR SPACE, IN HISTORY • POWERED BY: THE NEXT GREAT INDEPENDENT DATA COMPANY • MOVING INTO EUROPE & ASIA IN A BIG WAY, STARTING IN 2021 P A S T 5 0 V A S T D A T A 2 0 2 1 I T P T
  49. V A S T D A T A 2 0

    2 1 I T P T 5 1