Linux Can, Linux SAN
The company in question is NuTec Energy, serving the oil and gas industry with seismic imaging services. Based in Houston Texas, NuTec employs 35 staff. In early 2000, the company struck a deal with IBM to develop a massively parallel supercomputing system capable of dealing with the ever-increasing demands of seismic signal processing for the oil and gas industry-based applications.
The storage system initially consisted of 3000 Power 3/3+ CPU's with AIX on each server, and each CPU running its own analysis. The Network File System (NFS) file server utilized 2 IBM 'Shark' units connected to three B80 servers, and with shared file access to all CPU's. By 2003, however, the system was not keeping up with the demands being made on it, and so a project was established to specify a replacement.
According to Sampath Gajawada, manager of software development at NuTec Energy, "The target was a super-scalable SAN – a high-performance, single image storage environment using Intel, Linux, Fibre Channel and Ethernet." He defined several key objectives for the SAN:
- Software tuned to be latency-tolerant and massively parallel, buffered asynchronous communication & I/OThe main issues with the incumbent UNIX system were the high cost of the proprietary software and associated support and management, barely adequate computing power and bandwidth for some of the processing requirements, and a bottleneck on the storage NFS.
- High I/O bandwidth (>500 processing nodes)
- High computing power (processing power >2 Teraflops)
- Large flat file system (10-100TB), with easy storage management
- Cost effective, price/performance balance, scalable at low incremental costs.
"The existing system just couldn't cope with the demands of our Depth-domain Analysis and Time-domain Analysis," said Gajawada. "We had reached the stage where business requirements were forcing us to reconsider our entire system. We looked at all the alternatives, and settled on a combination of Intel and Linux."
Page 2: The SwitchThis decision to favor Intel/Linux enabled NuTec to create a lower cost structure with industry standard hardware, taking advantage of the improved FP (Floating Point) performance of Intel Pentium 4 processors, especially beneficial for their intensive graphical image processing requirements. The Linux route also eliminated the NFS bottleneck, and provided data sharing with SAN performance - with a CFS (cluster file system) on the SAN having the ability to scale to hundreds of nodes with minimal management.
NuTec adopted Minneapolis-based Sistina Software's GFS (Global File System) Linux cluster file system. Its cluster nodes physically share storage over fiber channel or shared SCSI, and while each node thinks the file system is local, file access is synchronized across the whole cluster.
In effect, GFS can pool storage onto cheap, efficient machines. NuTec's system resides on a Fibre Channel SAN infrastructure from LSI logic for high I/O performance. Processing consists of 350 dual processor P4 based nodes, providing 750 CPU's running on Linux, each one four times faster per box than the previous AIX processors.
The following table, prepared by NuTec, compares the two systems:
One of the main challenges NuTec experienced in the changeover was porting imaging software from UNIX to Linux. Though there were risks involved, the company saw it as an opportunity to reduce costs and management, and they made the transition in just four weeks.
As a result, definite cost savings have been achieved. The headlines are 50 percent fewer administrators and a 90 percent reduction in data center space needed, down from 10,000 to just 1,000 sq ft. "The bottom line is overall cost savings of 84 percent, including hardware and software," said Gajawada. "And, as a bonus, a higher adoption of Linux elsewhere in the company as a direct result of this implementation."