advanced_hpc_logo
phone
request_a_quote
Only Search Advanced HPC
       
       
 
   
   
  greenbulletHigh Performance Computing  
bar
 
  greenbullet Storage Solutions  
    bulletwhiteUnified Storage
 
    bulletwhiteParallel Storage Solutions
 
    bulletwhiteGPFS Solutions
 
    bulletwhiteObject Storage Solutions
 
bar
 
  greenbullet Cloud Infrastructure  
bar
 
  greenbulletHigh Performance Servers  
bar
 
  greenbulletData Storage  
bar
 
  greenbulletNetworking and Infrastructure  
bar
 
  greenbulletSupport  
bar
 
  greenbulletCompany Information  
bar
 
  greenbulletGovernment & Education  
   
   
 
GPFS Solutions
ParaStor GRS Scalable Storage Series
 
High speed, scalable, parallel storage solution for HPC, based on the IBM General Parallel File System

Advanced HPC has created the ParaStor GRS series of solutions utilizing GPFS (IBM’s high speed, parallel file system) along with tier-1 storage arrays to deliver a fully customizable, scalable high speed storage solution

Using the building block approach, we can deliver systems starting at 6GB/sec sustained performance. Need more performance or capacity or both, just add the appropriate building block. ParaStor easily scales with no disruption to service.

 
GPFS
   
The IBM General Parallel File System Removes Data-Related Bottlenecks
 
  • High-performance parallel file system
    - Stripes data across multiple nodes
    - All nodes access all data
  • Single global namespace
    - Existing apps get full performance
  • POSIX compliant
  • Supports a wide range of file system block sizes to match I/O requirements
  • Utilizes advanced algorithms to improve read-ahead and write-behind IO operations
  • Scalable distributed metadata and token management
    - All storage nodes perform metadata operations
 
     
General Parallel File System Simplifies Data Management at Scale
 
Two Advanced HPC ParaStor
GRS-5000 building blocks; two NSD servers, two controller pairs and two JBOD’s per building block. Total capacity of 1.9PB’s and 24GB/sec sustained performance. Mellanox FDR 56Gbps Infiniband infrastructure included.
GPFS provides online storage management, scalable access, and integrated management tools able to manage many petabytes of data and billions of files
 
  • Seamless capacity and performance scaling
  • Rolling upgrades of individual nodes in the cluster while the file system remains online
  • Quotas enable the administrator to manage file system usage by users and groups
  • SNMP interface allows monitoring by network management applications
  • GPFS supports POSIX and NFS V4 access control lists (ACLs)
General Parallel File System is Future Proof
ParaStor utilizes the IBM General Parallel File System to future proof your investment. GPFS adds analytics without adding a dedicated analytics infrastructure and fully supports IBM Big Insights Analytics. There is no need to purchase a dedicated infrastructure for analytics lowering your overall CAPEX and OPEX. Additionally, there is no need to move data in and out of a dedicated analytics silo helping to speed results. You receive enterprise-class protection and efficiency with full data lifecycle management.
 
Key Features of ParaStor GRS HPC
Scalable Storage Series
 
 
Scalable Building Block approach to    technical computing storage.
Two base configurations—GRS-1000    and GRS-5000; 6GB/sec and    12GB/sec, respectively.
Each Building Block offers redundant    NSD servers and two high capacity    storage shelves.
Enterprise grade SAS drives: 7.2k,    10k, SSD.
Capacity Optimized Building Blocks    from 240TB-960TB.
Up to 1.9PB per 42U rack.
IOPS optimized Building Blocks    capable of providing over 400,000    IOPS per Building Block.
Includes Mellanox High-Speed    Network Fabric for Infiniband and/or    10/40GbE.
 
 
 
GPFS Empowers Global Collaboration
 
Low latency access to data from anywhere in the world with Active File Management (AFM) distributed disk caching technology
 
  • Expands the GPFS global namespace across geographical distances
    - Caches local “copies” of data distributed to one or more GPFS clusters
    - Low latency “local” read and write performance
    - Automated namespace management
    - As data is written or modified at one location, all other locations see that same   data
    - Efficient data transfers over wide area network (WAN)
    - Works with unreliable, high latency connections
    - Speeds data access to collaborators and resources around the world
  • Accelerates schedules and improves productivity for globally distributed project
 
GPFS Manages the Full Data Lifecycle Cost Effectively
  • Policy-driven automation and tiered storage management
  • Match the cost of storage to the value of data
  • Storage pools create tiers of storage
    - High performance SSD
    - High Speed SAS Drives
    - High Capacity NL SAS Drives
  • Integrated with IBM Tivoli Storage Manager (TSM)
    - GPFS handles all metadata processing then hands the data to TSM for
      storage on tape
    - Data is retrieved from the external storage pool on demand, as a result of an   application opening a file for example
    - Policies move data from one pool to another without changing the file’s location
      in the directory structure
    - Thresholds can move files out of the high performance pool if more than 80%   full, for example
 

Building block approach—capacity increases as more building blocks are added to scale out the entire solution. If the solution is sold as standalone, one additional server node is required for quorum.

ParaStor GRS systems include a 3 year standard warranty to include advanced replacement of field serviceable units. Warranty can be upgraded to include next
business day or 4 hour onsite repair service*.

Solutions are integrated, cabled, delivered, installed and powered by Advanced HPC, dramatically reducing the time required to put systems into operation.

 
 
 
 



  Home | Products | Support | Company | News | PartnersPrivacy | Contact Us

Last updated 4/11/2016