Search the Community

Showing results for tags 'PCIe'.

More search options

  • Search By Tags

    Type tags separated by commas.
  • Search By Author

Content Type


  • Storage Hardware
    • SMB/Enterprise Storage
    • RAID Cards and HBAs
    • Solid State Drives (SSD)
    • Hard Disk Drives (HDD)
    • Home Storage and Computer Hardware
    • What Drive Should I Buy?
  • StorageReview News, Reviews and Projects
    • Storage News & Reviews
    • Enterprise Test Lab Reviews/Projects
  • General
    • Site Suggestions & Help
    • Storage and Tech Bargains
    • StorageReview Marketplace
    • The Bar & Grille
    • Reliability & Performance Drive Databases


  • Community Calendar

Found 4 results

  1. When I friend was looking for an SSD upgrade, I took advantage of it and sold my 'pedestrian' SATA SSD and replaced it with Samsung 950Pro connected via an adapter in a PCIEx16 slot. It took some wrangling but the drive is bootable on a Z87 board and is generally working fine. I ran some benchmark with Crystal DIsk Mark etc and they are more or less in line with reviews. However, when doing simple tasks, I'm not entirely impressed - for example I recently downloaded a zipped file - a compressed linux image. (650MB uncompressed) I wanted to unzip to the same folder and saw speed around 2-3MB/s, not exactly what I expected. The same goes for very small files - something that I'd expect to be instant. I check the activity monitor and CPU was hardly used, so clearly it is the SSD that's bottlenecking somehow? Finally I have 24GB RAM so I would expect Windows 10 to take advantage of that to cache recent files in RAM (such as the one I just downloaded)? Is there a way to investigate what's going on? I should probably add that while there is no radiator, the SSD is pretty well ventillated and running benchmark for longer indicates that there is no throttling even under have load for at least 10 minutes. The rest of the system: CPU i5 3.4GHz quad core 24GB DDR3 #1600Mhz c: - 256gb Samsung 950 Pro PCIE 3.0 x4 d: 4 way raid-0 - 4 x 500gb Sandisk Ultra II SATA SSD s: - 3TB backup HDD (nightly backups of d: and some folders on c: using CrashPlan - I temporarily turned it off, but no change) As you can see I have spare disks attached so I could create an image and perhaps reformat with a different stripe size perhaps?
  2. The OCZ RevoDrive 350 is a beast of an SSD with performance that will suit the needs of any enterprise, hardcore gamer, or power enthusiast. There simply aren't a lot great options when it comes to workstation PCIe storage but thankfully with OCZ re-committing, the RevoDrive 350 stands out as the De facto leader of the pack at its capacity points. OCZ RevoDrive 350 PCIe SSD Review
  3. Hello, I need to build a powerful Graphics-Workstation (Windows7-64 prof.) for Video editing. WS shall be equipped with one professional graphics controller (PCIe 16x) and 2x LSI MegaRAID-9270-8i controllers (both at PCIe 8x) reuslting in two RAID5-Arrays consisting of 8 drives each. The workload is as follows: Large files 2-6GB+ (sometimes several) are read from any of the 2 LSI MeagRAIDs while (often at the same time) Video-sequences are cut, pasted, rendered etc.. No other PCIe slots will be used. I find it hard to find a suitable motherboard that supports PCIe Version 3 without bandwidth limitation on any of the PCIe-lanes or slots. Which Motherboard(s) would conform to (or best meet) the setup described? Chipset Z87 supports PCIe Ver 3, but most boards I've seen multiplex the lanes with a PEX8747 chip. AFAIK Chipset X79 supports PCIe Ver 2. However, I would think the primary bottleneck might be the disk-write-speed (despite an 8-drive RAID5-array). Therefore a MoBo with PCIe Ver 2 would probably do the job as well. But which one? Any other thoughts or suggestions? Your advice is highly appreciated. Thanks
  4. We've started testing the OCZ ZD-XL SQL accelerator solution. We're using 16x900GB WD Xe HDDs in an iXSystems JBOD as a baseline and comparing those results against the 1.6 TB ZD-XL (Z-Drive R4 tuned for SQL caching). SQL testing is being driven by up to 20,000 virtual users in Benchmark Factory's TPC-C workload. We'll update this thread as we progress and are open to fielding questions or comments on the solution at large. More background information is available here -