A Revolutionary Architecture to manage all your data
WEKA is the modern file system that is uniquely built for those who solve big problems. by using a modern storage architecture to solve today’s data-intensive problems, eliminating the I/O bottlenecks that hold back innovation.
WEKA tackles the most extreme storage demands from the most data-intensive technical computing applications and software, including those in AI/ML, GPU acceleration, genomics, life sciences, financial analytic, and technical computing (HPC)
Epic Performance and Limitless Scale
The file system delivers record-setting performance for applications processing massive data sets, powering solutions to problems the world has never seen before.
Highlights:
- Built for the Cloud Seamlessly run on-premises, in the cloud, and burst between the two
- Shared Storage Architecture Enjoy shared access from any application to a single data namespace
- Optimized for NVMe Achieve lowest possible latency and highest performance
- Multi-Protocol Access Supports Linux, Windows, and native POSIX access to data
- Built-in Data Durability Uses distributed data protection and instant backup to S3 cloud for rapid recovery
- Advance Security Keeps your data completely safe with integrated encryption, key management, and access control
Workflows that unify artificial intelligence, data analytics, and modeling/simulation will make high-performance computing more essential and ubiquitous than ever. Chip ICT® HPC solutions are designed for maximum flexibility and ease of management.
Fields of Application:
Artificial Intelligence, Machine and Deep Learning
Different stages with the AI data pipelines have distinct storage requirements for massive ingest bandwidth, need mixed read/write handling and ultra-low latency, often resulting in storage silos for each state. This means business and IT leaders must reconsider how they architect their storage stacks.
- Reduces Epoch Times, While Delivering Lowest Inference Times
- Explain-ability and Reproducibility Using Snapshots
- Industry’s Best GPUDirect Performance, 113GB/sec for a single DGX-2 and 162GB/sec for a single DGX-A100
- In-Flight and At-Rest Encryption Delivers Data Compliance and Governance
- Enables Hybrid Workflows for Testing and Production
- More economical storage solution when running converged on the storage servers
High Performance Data Analytics (HPDA)
High Performance Data Analytics (HPDA) workloads are now applying predictive models for deeper insights, and success is measured by the fast execution of algorithmic and deep learning model results. HPDA is at the intersection of high-performance computing (HPC), traditional business analytics and artificial intelligence (AI). Critical applications, from fraud and anomaly detection, affinity marketing and business intelligence, are leveraging HPC technologies to gain competitive advantage. Weka is used by leading companies who are constantly seeking out new advanced technologies to gain a competitive advantage and improve business outcomes. HPDA requires a modern storage architecture that can handle the most demanding I/O intensive workloads and latency-sensitive applications at Peta-scale.
- Delivers 5x Improvement in Wall Clock Time for Long Running Projects
- Runs 10x More Workloads with the Same SAS Licensing. Eliminates Racks of Server Infrastructure.
- Best TCO Leveraging a Hybrid Storage Platform with NVME for Performance, and HDD for Capacity
- Enables Elastic Public Cloud for Peak Compute Demands
- Best Scalability, Up to Exabytes of Storage and Billions of Files in a Single Directory
- Single Global Namespace for Your Entire Dataset, no More Need for Multiple Data Copies
Life Sciences
Accelerate next-generation sequencing, microscopy or bio-imaging data pipelines, Lower the cost of research, and keep data secure
Affordable, next-generation sequencing, and bio-imaging backed by powerful analytics, offers the promise of huge advances in medical discovery and diagnostics. With data sets growing at an exponential rate, storing and analyzing mountains of data requires a modern approach for faster insights. High performance shared file storage is essential to transform the vast amounts of data into meaningful discovery. The first flash-native parallel file system, WekaFS is ideal for accelerating the data pipeline in Life Sciences. Weka eliminates performance bottlenecks and significantly reduces the complexity and cost of managing high-performance storage systems at scale.
- Speed Up Analytics By 10x With Low Latency Flash Performance
- Improve Collaboration with Peta-scale Single Global Namespace
- Stretch Funding Further with Integrated Tiering to Low Cost S3 Object Storage
- Protect Precious Patient Data with Secure Encryption and Automatic Backup
- Manage Peak Demand Periods with Cloud-Bursting to AWS for Elastic Compute
- Best support for GPU compute to handle CryO-EM workload
Financial Services
Dramatically reduce time to trade with a high bandwidth low latency data platform
Companies are now applying predictive models for deeper insights and risk mitigation, and success is measured by the fast execution of algorithmic and deep learning model results. WEKA is used by leading quantitative trading companies who are constantly seeking out new advanced technologies to gain a competitive advantage and improve trading outcomes. Financial analytics requires a modern storage architecture that can handle the most demanding I/O intensive workloads and latency-sensitive applications at Peta-scale. WEKA has demonstrated record-breaking performance on STAC M3 benchmarks using kdb+ time-series database.
- Delivers Higher Performance Than Local-Drive SSDs
- Enables Elastic Public Cloud for Peak Compute Demand
- Single Storage Platform for Entire Dataset
- Dramatically Improves Wall Clock Time
- On-Demand Scalability and Economy
- Best TCO Leveraging NVMe for Performance and HDD for Capacity
Building Blocks
AMD with PCI-E 4.0 | Intel with PCI-E 4.0 | |||
Form Factor | 8U | 16U | 4U | 8U |
Usable Capacity | 363TB | 726TB | 218TB | 436TB |
Storage Media NVMe* | 80x(10x 7.68TB/node) NVMe Gen4 |
160x (20x 7.68TB/node) NVMe Gen4 |
48x (6x 7.68TB/node) NVMe Gen4 |
96x (12x 7.68TB/node) NVMe Gen4 |
CPU | 8x AMD EPYC™ 7402P Processors** | 8x AMD EPYC™ 7402P Processors** | 16x Intel® Xeon® Silver 4314 Processors | 16x Intel® Xeon® Silver 4314 Processors |
RAM | 2TB | 2TB | 2TB | 2TB |
Network | 16x single-port Mellanox CX-6 200G VPI |
16x single-port Mellanox CX-6 200G VPI |
16x single-port Mellanox CX-6 200G VPI |
16x single-port Mellanox CX-6 200G VPI |
IOPS*** | 10.2M (1.275M per node) |
8.4M (1.05M per node) |
8.5M (1.06M per node) |
11.28M (1.41M per node) |
Bandwidth*** | 235GB/s (29.4GB/s per node) |
280GB/s (35GB/s per node) |
202GB/s (25GB/s per node) |
286GB/s (35.8GB/s per node) |
I/O Latency | < 200μs | < 200μs | < 200μs | < 200μs |
SW Subscription | 1yr, 3yrs, 5yrs | 1yr, 3yrs, 5yrs | 1yr, 3yrs, 5yrs | 1yr, 3yrs, 5yrs |
* To request drives of a different capacity, please contact our sales team
** 3rd Gen AMD EPYC™ 7003 Processors are available.
*** Estimated; number varies based on actual components used.