6 ms·
Sequential read/write is usually optimized for burst workloads on consumer ssd's. The Intel 900p can sustain its peak read/write performance for hours. Anyway,
by olavgg 8y ago
Sequential read/write is usually optimized for burst workloads on consumer ssd's. The Intel 900p can sustain its peak read/write performance for hours.
Anyway, the big difference with Optane is that its performance is more similar to having more system memory. I have not benchmarked this, but I believe a system with just 8GB ram + Optane will run better than a system with 16GB ram and Samsung 970 pro for workloads using up to 16GB of ram. There is a reason Intel announced its memory drive technology for Optane DC4800x https://www.intel.com/content/www/us/en/software/intel-memory-drive-technology.html https://www.intel.com/content/www/us/en/software/intel-memor...
EDIT: adding a link to lwn.net article explaing the use of swap/memory overcommit https://lwn.net/Articles/704478/ https://lwn.net/Articles/704478/
- wtallis 8y ago> but I believe a system with just 8GB ram + Optane will run better than a system with 16GB ram and Samsung 970 pro for workloads using up to 16GB of ram. No, not even close. PCIe is still way slower than DDR4. Once your working set grows beyond available RAM, the Optane SSD will be a better swap device than any flash-based SSD, but you still notice your system slowing down drastically from all the swap activity. Intel's Memory Drive Technology is really intended for situations where you have multiple Optane SSDs, and a workload that wants a large amount of total RAM but seldom has a true working set larger than actual DRAM capacity.
- eloff 8y agoNothing will beat a workload requiring 16GB of RAM, than a system with 16GB+ of RAM. You don't even need to benchmark to know that much.
- walrus01 8y agofor comparison, write speed to RAM on DDR4-3200 is somewhere between 50000 to 60000MB/second on a fairly normal skylake workstation platform.
- dimensi0nal 8y agoPCIe latency is probably the bigger issue.
- dogma1138 8y agoThis. Which is why Intel is working on Optane DIMMs. Traversing the root complex can take upto 100 cpu cycles each way.
- Makepio 8y agoyep but Optane DIMM /NVDIMM is not for client product. It is on data center product list. Persistent Memory isn't just a concept on the textbook. It be realized now.
- dogma1138 8y agoSo was Optane at first and so were SSDs in general. DDR4 already has an optional spec for NVDIMMs, DDR5 will make it mandatory.
- x86awesome 8y agoPretty exciting news. Dudes can't wait to buy NVDIMM and plug them into their desktops.
- yyyr 8y agoVery exciting news. People cant wait to buy NVDIMMs and then plug them into their desktops.
- Makepio 8y agoPlease forget Optane, just think about intel persistent memory. If you really care about performance /latency.
- lorenzhs 8y agoNope that’s more like cache write speed. Single-threaded sequential access to RAM is typically more like 10-15GB/s for current server CPUs: https://panthema.net/2013/pmbw/results.html https://panthema.net/2013/pmbw/results.html - it’s true that workstation machines tend to have somewhat higher sequential throughout compared to the server-heavy list at that URL, but it’s more like 20GB/s, not 50-60.
- DSingularity 8y agoOr a system with faster RAM or faster swap device :-D