I have been using the Ultime Boot Disk for the past few months to test x86 and X64 hardware. The disk contains numerous awesome utilities that can be used to test memory, disks and CPUs. The following packages come on the CD, and are four of my personal favorites:
- Memtest86+ to test memory
- Darik’s Boot and Nuke to securely erase data from a disk drive
- CPU burn to test CPUs
- PCI sniffer to identify the type of card in a system
If you are curious what the various UltraDMA modes mean, and the speeds they operate at, you might be interested in the bugclub UltraDMA tutorial.
If you are interested in learning why the AMD Opteron processor is currently the top dog in the X64 server space, you might be interested in the following Anandtech article:
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=1815&p=1
The article discusses the K8 architecture in depth, and explains why hyper transport rocks. I have been wanting to upgrade my Sun Ultra 10s to more modern hardware, and I think an AMD solution would be a good fit (especially since Solaris FMA now supports AMD CPUs!).
This is a nifty write up on AMD Opteron frequency scaling and the issues associated with TSC drift:
While reading through some Fujitsu documentation I came across the following:
“Soft errors were formerly caused by alpha particles in the IC housing. This can now be precluded for the most part by an improved IC housing. Radioactive gamma radiation from solar eruptions is now the main cause of soft errors. In all probability, pure soft errors do not lead to uncorrectable multi-bit errors in a memory chip. Soft errors tend to occur by chance and can be eliminated for the most part by ECC. ECC memory protection is now a standard means of ensuring data security in servers. Single errors in more than 2 memory ICs can be neglected, since the likelihood of faults at identical addresses and at the same time is very low.”
This leads me to wonder if data centers can be partially shielded from gamma radiation?
If you are looking to learn more about PCI express, you will definitely benefit from looking at hardware secret’s PCI Express tutorial! They provide a good introduction to PCI express theory, and provide numerous charts to show bandwidth and available configurations! Good stuff!
While ignoring my own advice and reading various Mac rumor sites today, I saw that the next generation Powermacs may ship with DDR2 memory and PCI express video cards. Being the geek I am, I wandered off to see how much throughput could be achieved with each technology.
I started my knowledge quest by reading PCSTATS PCI Express tutorial, and was blown away by the potential of PCI Express. Not only does PCI express move considerably more data than PCI (250MB/s vs. 133 MB/s), it uses serial switched lanes to move data, and allows devices on the same PCI express BUS to communicate directly without involving the chipset.
Now DDR2 doesn’t look as promising as PCI express, and requires the memory to run at a high clock rate to achieve higher throughput than good old DDR memory. I am always curious to see what Apple does, and look forward to reviewing the specifications of the next generation powermacs!
While reading through Mike Shapiro’s FMA presentation today, I came across two cool new hardware technologies. The first is FBDIMM, which Micron describes as:
Address/command soft errors can disrupt server performance and reliability. To help lessen their occurrence, Micron’s FBDIMMs incorporate an enhanced cyclic redundancy check (CRC) that provides greater data and address/command protection than traditional server modules.
Designers can also configure it to suit their particular applications. Providing an even greater defense, the bit lane fail-over correction feature identifies bad data paths and removes them from the operation. Together, these error detection methods dramatically reduce address/command soft errors.
The second technology is CPU chipkill, which Findany ISP describes as:
CHIPKILL - A technology developed by IBM for servers and other systems that demand high availability. It allows a computer motherboard and BIOS to detect problems with the computer’s memory and selectively disable problematic parts of the memory. Depending on the technology used, this technology may or may not require specialized memory chips.
Hopefully Fujitsu and Sun will integrate these technologies into their next generation APL server line.