Search results

I'm going to have to politely beg to differ on these points. I have observed the H240 to have subtle but fatal flaws which render it incompatible with the vast majority of SATA SSDs I have tested when behind the HP SAS expander part. The issue presents only under heavy and constant I/O...

Based on 'cable mellanox' I would guess possibly the supported transceiver restriction. You should see something in dmesg about ports being disabled due to unsupported transceivers if you're binding i40e. If you're binding all PFs to DPDK, it may fail to establish a link with no indication...

I think it's worth mentioning here that while some of the dual port 40GbE chipsets are limited in a real way by PCIe 3.0 x8 the XL710 is not. I've measured peak throughput of ~56Gbps on dual port ConnectX-3 cabled as 2*40GbE in LACP and suspect it may be capable of more than that given an ideal...

I confirmed continuity between pins 1-2-GND and continuity between pin 3 and one of the power MOSFET terminals, and pin 4 must be 5V from earlier analysis, so the power connector is definitely GND/GND/12V/5V as stated earlier. I have not powered one on yet and probably won't bother until I...

I would recommend the following topology with the parts you have on hand: From the 24 port expander, 4 ports for cross connection to other expander, 4 ports for upstream, 16 ports for drives From the 36 port expander. 4 ports for cross connection to other expander, 4 ports for upstream, 20...

At the protocol level there's very little difference between the 4 lane wide link and the 8 lane wide link, or for that matter 2,3, 5, 6 or 7 lane wide links. If it generically supports 8 lane wide ports at all it "could" support wide ports facing anything including other downstream expanders...

Jeez that was hard to find. Apparently the MD1280 was called the Compellent SC280 previously. It was also available from Sea Micro, Xyratex, and Open Storage Solutions as the SP-2584. Crazy. Being such a niche item I think it provides an unusually transparent view into the wild world of...

The LSI SAS2x36 in general does not give a **** where you connect things as it is a full SAS switch capable of supporting many topologies, although I haven't bothered to experiment with an Intel branded one and they theoretically could have limited it with custom firmware. There is a difference...

So I did a little more visual analysis of the board layout to try to find an identifiable interface or chip that I could reference to a voltage specification to confirm the pinout. I noticed pin 4 of the power connector goes directly to what should be pin 2 VCC on the black connector labeled...

Unfortunately yours is probably toast if that theory is true. At least the replacement will be cheap, eh? ;)

I stumbled across these recently on eBay and shortly afterwards located this thread while researching them. I believe these are the SAS expander internal to the EMM of the MD1280. It is the only Dell storage chassis I have been able to locate with a similar spacing of the external SAS...

I do have 1500 Pros and I can provide data from them but I'm afraid I do not have Windows available. What metrics are you interested in or do you have e.g. fio config I can run?

I have somewhere in the ballpark of 100 assorted Intel consumer and enterprise SSDs in my lab and have tested hundreds of drives from pretty much every major manufacturer. While it is true that many low end consumer SSDs are unable to sustain their specified performance for very long, it's...

Erm.. as I stated in the post you quoted, the Intel 1500 Pro has 4x the specified write performance and virtually the same specs in every other way as the S3610, except at 20x lower specified power. The S3610 manages not much more than 10% advantage over the 1500 Pro in any benchmark for its...

Does anyone know why the power consumption on certain (I want to say recent but honestly have no idea what the pattern is) SSDs is so high? The elderly Intel 1500 Pro 480GB is specced 125mW idle 195mW active. The S3610 480GB is specced at 600mW idle and 4.3W active, which makes the idle power...

The big deal with cut through in storage applications is that you should be using jumbo frames in pretty much any storage application. Even if your hardware is capable of receiving at line rate it will significantly reduce the CPU load of processing that flow. That said, it is virtually...

Quite likely, considering I've benched >8GB/sec aggregate multi-node sequential write throughput across my Ceph cluster of 4 node * 6x OSD of mixed Intel 520 (800mW peak write) and 1500 Pro (200mW peak write), 1x Optane 900P per node shared by all its OSDs as blockdb/WAL (14W peak write, but at...

I keep looking at these and seeing that there are still some in stock then seeing the nearly 7 watts on load figure and remembering why I noped on them. That's >50x more than a decent SATA SSD which can do >500MB/sec read and write. In a 24x2.5 that would be 160W of additional power...

I do believe you are. Elsewhere on Intel's site it's specified as 70GB per day. That's pretty comical. It would take 58 seconds of sequential writes per day to reach that figure or around 30 total hours of sequential write to reach the lifetime write specification. I would say very...

This switch supports Ethernet mode operation only. It does support RoCE (RDMA over Converged Ethernet), which is more or less Infiniband over Ethernet. Pretty much anything that supports DCB (datacenter bridging) or PFC (priority flow control) supports RDMA, though, and technically speaking...