INSIGHTS
What are High Frequency Servers?
Speed, execution and money all go hand in hand. Time to complete a transaction is imperative within the financial world and this is where High Frequency Servers come into play.
Creating Alerts to Detect Critical Hardware Errors
The main aim when monitoring key elements of your server infrastructure is to create alerts, one such example is e-mail alerts. When a hardware error is discovered in the Supermicro IPMI event log, there are many ways of implementing such alerts. You can script this...
Power Hungry CPUs and “Phantom” Throttling Featuring the 10980XE
Intel’s 10980XE can require more than 500+W to reliably run at overlocked speeds surpassing ~4.7GHz. In our testing we found an issue that was more difficult to spot with the 7980 or 9980. At that time we could hit what we saw as the maximum safe daily overclock on...
Intel Software RAID vs. Hardware RAID
This post will feature both a general overview of Intel software RAID vs. hardware RAID. The comparison will include performance numbers. PROS CONS No additional hardware required RAID levels are limited Cheaper cost No dedicated cache Very easy to use Slower...
Supercomputing & High Density Racks
At ICC we’ve now deployed the ICC AXIS™ R-725a. Testing on the EPYC 7000 series in our lab showed an increase of ~.1GHz in performance per core on these high core count processors with liquid cooling vs. air. While that does not seem like a big number, it yields a...
ICC COVID-19 Response
To maximize the containment of COVID-19, the state of Illinois has issued a state-wide “stay at home” order as of Saturday, March 21, 2020. While the evolving COVID-19 situation has presented new challenges, ICC is well prepared to continue to respond with the best...
Prepping a System for Subzero Overclocking
In overclocking there is a constant issue of heat created by the CPU. Not only does it lower the stability of a processor as the temperature increases it also can cause damage to the CPU and lowers the tolerance to higher voltages which further causes damage....
Linux System Load Averages
Helping to Understand the Mystery While load averages are a metric that are widely used for monitoring the health of a system. Quite a few organizations use this metric for capacity planing without fully understanding where the limits should be. But most admins are...
Integrating Windows Deployment Services with existing Linux based PXE server for customer imaging needs
IT infrastructure in today’s world is far from black and white. It needs to be dynamic and have many different types of technologies mesh together to fit ever-evolving customer and user needs. In our case, ICC uses a PXELINUX deployment to serve different types of...
The Super-binned 9990XE Perfectly Illustrates Current and Future Issues with Power Draw and Cooling
The 9990XE made headlines in April of this year for being an Auction-Only part as well as Intel's fastest most powerful CPU to date. At International Computer Concepts we've tested a few of these at their Manufacturer's settings (5.0GHz, Default voltages) and found...
Broadwell-E CPU Benchmarks
Broadwell-E has been released a few weeks ago to consumers, and we have had our hands on some samples previous to that. We would like to share some of the relevant specifications of the BW-E family and detail some of the early results we have seen with them....
ICC Blackhole HPC Workstations
High performance computing can be defined as aggregating computing power in a way that it can solve large scale problems in various fields such as science, business, etc. These can range from the Tianhe-2, the Chinese “Milky Way” supercomputer which has been built to...