Tag Info

Hot answers tagged

19

Note: This answer is specific to the server components described in the OP's comment. Compatibility is going to dictate everything here. Dell PERC array controllers are LSI devices. So anything that works on an LSI controller should be okay. Your ability to monitor the health of your RAID array is paramount. Since this is Dell, ensure you have the ...


8

Yes, the SSDs will be way faster than the SAS drives. For sequential throughput, a good RAID of SAS drives might do pretty well, but for random access, the SSDs will blow them out of the water which can result in a very noticeable performance difference. Depending on the particular SAS drives and the particular SSD drives, the SSDs may have a better ...


8

You have one drive in a RAID that is misbehaving, and producing occasional errors? Sounds like a hardware problem, and one that is likely to get worse. You should consider replacing the drive. Yes, it's expensive, but how much is your time worth, and how bad would it be if the entire drive went south at an inopportune moment?


8

What you have probably heard is that for RAID5 configurations using large disks problems can arise under some circumstances. These problems are not an issue for RAID 1 as the rebuild I/O is less than that of RAID5. What's the recommended size? - that which will do the job with some elbow room for expansion. we will be running daily backup so we can just ...


7

Consumer grade SSDs will work fine in many servers for use cases. They are way, way faster then SAS disks. I'd suggest the reason to get enterprise disks over consumer disks is not the speed, its the read-write cycles and better engineering - for example supercaps are present in some enterprise SSD's where the consumer grade version does not have this - ...


6

If you are using them for writes, to avoid data corruption in the event of power failure you need to make sure that you only consider models with a supercap. Eg. Intel S3500, Samsung 845DC Pro Otherwise consumer SSDs are more suited to caching.


5

The performance inconsistency of consumer SSDs can cause problems with some raid controllers, the spikes in I/O latency are exacerbated when using a raid controller as it often will not be using TRIM (I don't know of any controller that does). Enterprise drives are designed around consistent performance even without TRIM so they typically play well with RAID ...


5

The reason to go with enterprise grade gear is reliability more than speed. Most consumer SSDs are MLC, with the lower end stuff being TLC (MLC does 2 bits a cell, TLC does 3, and they're less performant, and reliable than SLC). At some point, they may also drop the onboard ram cache to save costs, as nand cells get cheaper. A enterprise SSD also has greater ...


4

Even consumer-grade SSDs are much faster than the faster 15k HDDs, so from a performance standpoint they will be fine (if using the right disk and if overprovisioning them), but you had to carefully pick them, especially due to how they interact with hardware-based RAID controller... First, check if affordable, entry-level enterprise grade drive (as Intel ...


3

Can you see the SMART information for the individual disks via the Disk Utility? Look at the Pre-Fail checks & see if anything stands out. This will indicate a failing physical device. Is the disk that's producing the errors in a RAID group with other disks? If not it may be the file system and not the drive or, depending on the layout, could be the ...


3

Fairly new to Azure, but here's my offering: The more Azure disks you span the better your performance will be, this is true. I'm not sure about the equation though. Either way, I think you're barking up the wrong tree... Azure imposes artificial limits on metrics such as IOPS and Max reads/writes per second, so the queue length is only one of a few ...


3

There is no particular limit on how much RAM a 32-bit process can use. A 32-bit process can only directly map 4GB of virtual memory at a time, but that's a virtual memory limit, not a RAM limit. For example, a 32-bit process on a 64-bit machine can access a 32GB file that is cached entirely in RAM, thus reading to, and writing from, 32GB of RAM. This is ...


3

I don't think it's worth using this server in ANY situation right now. Even an entry-level modern $500 system will be more capable and robust. I'm not sure I understand the desire to use something for "non-critical functions" just because it's there. To answer your specific questions, though, Ultra 320 SCSI drives topped out at 300GB 15k RPM before SAS/SATA ...


2

Probably not. You've added a foreign component to the HP system. It can't be expected to work the same way as native parts. My experience with LSI controllers running the internal disks of a ProLiant server is that the disk LED activity is random or may not correlate to what's actually happening. But in the end, it really doesn't matter. It's worth noting ...


2

NONE of those options are valid, you can do it in two ways; RAID 10 - you'll need 80 SSDs of that size to give you 100TB, though that doesn't include the typical overhead you'd see so I'd personally go for 90-100. RAID 60 - you'll need 44 SSDs of that size to give you 100TB, this is via 2 x 20+2 R6 arrays, again I'd actually increase this to 50-60 to deal ...


2

it is not good idea to store your backup data on the same disk, because if the disk fails someday, you will lost the data. I will suggest you to keep your backups on the separate disk, cloud or tape solution. This way you will prevent lost important data. As for the disk, I will suggest you to calculate how much disk space you will need with eventual growth ...


2

4 GB is the virtual limit. 2 GB is the physical memory limit for 32-bit processes, unless the application is linked with the "LARGEADDRESSAWARE" flag. This sets the 0x0020 bit in the characteristics section of the PE header. But the application may also launch multiple processes in the course of normal operation, each of which may consume it's own memory, ...


2

You're focusing a bit much on a single metric! Diagnosing a bottleneck to a single component rarely ends with one counter giving a full explanation. There are quite a few great guides for using perfmon to diagnose performance problems on SQL Server. And unfortunately your Admin could be right, the counter you choose does indeed depend on the underlying ...


2

Not using a partition is totally fine. The only struggle that can appear is, if you're exchanging a disk with a different vendor and your new disk is a few blocks smaller.


2

Similar to this issue, you need to have a cache memory module on your RAID controller in order to have more than 2 logical drives on a Smart Array P410 setup. See: HP ProLiant DL360p Gen8 Server - Unable to Create More Than 2 Logical Drives


1

Sorry haven't experienced the same, but we run LSI and have had firmware updates smooth things out before. Please check that you have the latest firmware for the device.


1

The better start can be the set of following rules: disable logging and accept_mutex enable sendfile set sendfile_max_chunk Configuration: events { accept_mutex off; } access_log off; sendfile on; sendfile_max_chunk 512k; New Nginx (1.7.11 or newer) feature thread pool can be really helpful in your case: location / { root /home; aio ...


1

The devices should show up automatically in Linux under VMware. Check the output of dmesg|tail. If you've changed the size of the devices, you can rescan/recognize this with... echo 1 > /sys/class/scsi_disk/0\:0\:0\:0/device/rescan, where you substitute the SCSI disk ID. For example: [root@xt /sys/class/scsi_disk]# ls -1 2:0:0:0 2:0:1:0 3:0:0:4


1

Write cache: disabled. Do you want the write cache disabled? If not, execute hdparm -W1 /dev/sdb and verify it was changed with hdparm -W /dev/sdb. Rerun your test. If it looks better add write_cache = on to /etc/hdparm.conf.


1

From both the server spec sheet and the SmartArray 6i manual, the maximum HDD size is capped at 300 GB. While I think that this is not an hard limit, the fact that you had to use old Ultra320 disks (not newer SAS ones), let me think that you can not buy bigger disk at all. Regarding the supported disks, from my understanding HP controller have no problem in ...


1

You are proceeding in the wrong order. Please follow these steps: first, BACKUP YOUR DATA (and take a snapshot) use GParted to resize /dev/vda2 first and /dev/vda5 after reboot your virtual machine use pvresize /dev/vda5 and lvresize to resize your physical and logical volumes, respectively. EDIT: how to resize partition using fdisk To resize your ...


1

Your operation plan is OK. The only addition to it is that after having removed the disk and rebooted the machine, you will have a disk in the "failed" state. To completely remove any reference to such disk, you had to issue the following command: mdadm <mdarray> --remove <faileddisk> Obviously, substitute and with the read device names. A ...


1

Why do you want to put one pair of huge disks in there, when you have a server running a database? If you want to look at performance, try putting your database on different disks than your OS / web application. Databases are often bottle-necked on I/O. In stead of purchasing 2x 2TB disks, you could buy 2x 320GB disks for the OS / web app, and 2x 128GB SSD ...


1

There should be tools available to query your RAID controller and determine the SMART status of the drives in the array. Not knowing the particular device you've got, I don't have any suggestions as to what to use. Once you know what to use (and how to use it), you'll need to automate the monitoring, so it will proactively notify you when there is a ...



Only top voted, non community-wiki answers of a minimum length are eligible