This article provides in-depth details of the NVIDIA Tesla K-series GPU accelerators (codenamed “Kepler”). “Kepler” GPUs improve upon the previous-generation “Fermi” architecture. For more information on other Tesla GPU architectures, please refer to: Important changes available in…
This article provides in-depth discussion and analysis of the 22nm Xeon E5-2600v2 series processors (formerly codenamed “Ivy Bridge”). “Ivy Bridge” processors improve upon the previous 32nm “Sandy Bridge” microarchitecture and are available for sale as of September…
Ethernet Performance of ethernet networks varies widely. Factors include the switch and NIC manufacturer, firmware settings and system/software settings. Physical layer even plays a role: 10 GigE over RJ45 (10GBase-T) has a higher latency than SFP+ Direct-Attach…
Transistor density doubles every ~2 years. Ideally, this means we can provide you computers that contain twice as much memory and perform twice as fast. In practice, it’s much more nuanced. Benchmark Your Application The best performance…
This article applies to groups of computers which consume 1,000+ Watts of electricity. Review your quote or contact us to determine how much your systems will require (the power and cooling numbers are typically listed at the…
The following items should be completed to maintain the health of your Linux cluster. For servers and workstations, please see Common Maintenance Tasks (Workstations and Servers). Backup non-replaceable data Remember that RAID is not a replacement for…
Whether you’re working on a cluster, a server or a workstation, most installations of Linux are similar. When something goes wrong, you need to determine the exact issue before you can get it resolved. This article provides…
Best Practices for managing servers with IPMI Baseboard Management Controllers (BMC) with IPMI are commonly used to manage servers remotely. Almost 100% of Microway’s servers support IPMI either through a dedicated management port or a shared LAN…
Memory The following values are measured per CPU socket. They must be doubled or quadrupled to calculate the total memory bandwidth of a multiprocessor workstation or server. For dual-processor systems, multiply by two. For quad-processor systems, multiply…
The following items should be completed to maintain the health of your workstation or server. For compute clusters, please see Common Maintenance Tasks (Clusters). Backup non-replaceable data Remember that RAID is not a replacement for backups. If…
