Inspur HPC + Big Data Solution

Solution Briefing

With the increasing data demand, mainstream big data processing architecture – Hadoop has been facing the bottleneck of computing and interconnects capability, which is the advantage of HPC technology. Therefore, more and more end users are looking for the convergence of Big Data and HPC with a unified infrastructure and management software.

Inspur HPC + Big Data Solution manages, monitors and schedules common HPC and Big Data workload. We can combine traditional HPC workloads with Hadoop workload through the same data accessing by HPC application and big data application.

"We have made it easy to stand up a complete HPC cluster with the Big Data and manage it in the same way as you manage a cluster"

Inspur HPC + Big Data Solution

Solution for Customer Who Seek to:

• Unify operation interface

HPC & Hadoop unified interface
- Job management for HPC/Hadoop
- Dynamic scheduling for HPC/Hadoop
- Provisioning for HPC/Hadoop
- Unified resource charging system

• Unify hardware system management
- HPC/Hadoop resource monitoring
- Dynamic switching between HPC/Hadoop nodes

• Unified storage system, data copy free
- IEEL files system Hadoop connector
Apache Hadoop on Lustre Connector eliminates the need to copy data to the Hadoop Distributed File System (HDFS). The Hadoop on Lustre Connector allows both Hadoop and HPC Lustre clusters to use exactly the same data without having to move data between file systems or storage devices. Hadoop ecosystem tools like Mahout, Hive and Pig are now able to utilize the Lustre File System

Inspur HPC + Big Data Solution delivers dynamic scheduling, provisioning, and management of multi-step/multi-application services across HPC and Big Data environment.


HPC+Bigdata solution.pdf

Home Products Industry Application Partners News About us
©1996 - 2014 INSPUR Co., Ltd. 鲁ICP备05019369号