Geekly Articles each Day
📜 ⬆️ ⬇️

How to choose hard drives for servers?



There are many myths in the IT area. “You can unsubscribe from spam,” “Two antiviruses are better than one,” “Server hard drives should only be branded.” When replacing and expanding the railway fleet, there are a lot of nuances and subtleties that need to be taken into account, and this is not without its prejudices. What are the hard disk drives for servers, how they differ, what you need to pay attention to, and whether they should be with the server manufacturer's logo - read about it under the cut.

If the disk is installed in the server, then it must meet strict requirements for:


In other words, the hard disk in the server should be like a pioneer - always ready to handle multiple requests with minimal latency, ensuring a high level of data integrity. In high-load servers, hard drives have been working intensively and non-stop for years.
')
There are four main categories (we do not take into account SSD, SAS SSD, PCI-e SSD) hard drives:


First of all, you need to decide on the connection interface - SATA or SAS.

SATA or SAS?




The SATA interface is a development of the IDE, which was later renamed PATA . That is, this interface is initially focused on use in consumer computers, as well as in industrial systems with moderate requirements for performance and reliability. At the same time, SAS is a successor to the classic "server" SCSI interface.

Initially, the SAS interface had a higher bandwidth than SATA. But progress does not stand still, and the third generation SATA III has a maximum bandwidth of 6 Gbit / s, as the second generation SAS. However, servers with a third-generation SAS controller are already available on the market, with a throughput of up to 12 Gbit / s.

To connect SAS-drives server must be equipped with the appropriate controller. At the same time, the interfaces are backward compatible: you can connect SATA disks to the SAS controller, and vice versa — you cannot.

SAS provides full-duplex data exchange: a hard disk processes one command at a time for reading and writing, and a SATA disk for either reading or writing. But this advantage will be noticeable only with a large number of disks, if we compare SAS NL and SATA RE.

To sum up the intermediate result: SATA drives are good for creating large-scale storage, which do not require maximum performance. And if you need to squeeze everything out of the disk subsystem, then your choice is SAS.

Say no to regular hard drives.




Immediately clarify - the usual SATA is not intended for use in servers. There are several reasons for this:


Of course, the cost of the usual desktop SATA is significantly lower than that of the server, and nothing prevents to use them for small tasks that do not require high performance of the disk subsystem. If the safety and speed of access to data is in the first place, it is still highly recommended to take server hard disk drives.

Vibration Resistance


To solve more or less demanding tasks, it makes no sense to put only one disk. To ensure the minimum level of reliability of data storage, you need at least two drives combined in a RAID. But when 4 or more devices are collected in the basket, the vibrations arising from their operation affect the stability of the spindle rotation and the accuracy of the positioning of the heads. Therefore, server hard drives have a number of constructive differences from household:


Household disks of all this are deprived. With a sufficiently strong level of vibration, the likelihood of read / write errors for ordinary SATA is 50% higher than that of SATA RE .

Unrecoverable Error Level


The next difference between server hard drives and home drives is the level of unrecoverable errors. With conventional SATA, it is about 10 -14 (1 bit for every 10 14 bits = 12.5 terabytes). That is, with a six-fold rewriting of a double-byte disk, you will almost certainly get one unrecoverable error. For household drives, this is not a problem. But if you rewrite the database every month, then after six months it may turn out to be a bat.

=
(N * (X / 12500 * 12500) / 12500) * 100%

Where:


Suppose you have created a RAID 5 array of 1 terabyte normal SATA. When you rebuild the array, you will get an unrecoverable error with a probability of 40%.

=
(5 * (1000 / 12500 * 12500) / 12500) * 100% = 40%.

And if you are using 600 GB disks, then the probability of epic fail with rebuild is 24%:

=
(5 * (600 / 12500 * 12500) / 12500) * 100% = 24%.

In SATA RE and SAS NL, the level of unrecoverable errors is 10 -15 , that is, an order of magnitude less than in conventional SATA. Then in our example with RAID 5 we get:

For 1 terabyte disks, the probability of error is 4%.
For 600 GB drives, the probability of error is 2.4%.

For SAS disks, the level of unrecoverable errors is even lower - 10 -16 :

For 1-terabyte disks, the probability of error = 0.4%.
For 600 GB disks, the probability of error is 0.24%.

Please note : the probability of an error is proportional to the number of disks in the RAID.



How does a lower level of errors occur in SAS disks? Magic .


Work in a RAID


Another important drawback of conventional SATA is the lack of error recovery when working in a RAID array. Suppose you hope to backup, and for the sake of saving built a RAID from the usual SATA. When an error occurs, the hard disk repeatedly tries to read the failed block. And while he is doing this, he is not responding to the signals from the RAID controller. He perceives it as a hard disk failure, excludes him from the array and tries to recover. In other words, when an error occurs, the entire disk drops out of the array.

In the case of SATA RE, SAS NL and SAS, the situation will evolve differently. Upon detecting an error, the disk informs the controller about the presence of a bad block. The controller requests this block from other disks in the array and transfers it to the failed disk. At the same time, the device does not drop out of the array, and there is no performance drop.

Brand Myth


Finally, the most important question: do you need to buy “native” disks?

It's no secret that HP, IBM and DELL do not produce hard drives. They buy them from third-party manufacturers, after which they test, reflash and glue their logos.



On the one hand, such discs have several advantages:


But for all the good you have to pay - "native" hard drives sell about two to three times more expensive than the same models, but with the logos of manufacturers - Seagate, Western Digital, Toshiba, HGST.

As you understand, this price difference is far from being justified by the promises of increased reliability. Therefore, probably, many have heard that non-native hard drives are working in HP, IBM and DELL servers unstable or too slow. Someone even scared that the server will not start with the “left” disks.

Where do these claims grow from?

The vast majority of servers use disk subsystem performance improvement technologies. It is for this purpose that vendors reflash hard drives - to provide support for these technologies. If you put "non-native" discs, then you just can not use the company know-how, nothing more.

Also earlier, vendors artificially forced to use “proprietary” drives, using locks at the controller level. In the end, the anger of the masses forced over time to abandon this vicious practice. Today problems most often arise with relatively old server models. And this is solved by simply uploading the latest firmware to the controller. Although there are simply capricious models of controllers, for example, P410 in HP servers.

As practice shows, "non-native" hard drives work without difficulty:


Non-native hard drives are fully compatible with both server sleds and internal monitoring systems. But the sled needs to be set only for relatives, and only for a certain generation of the server. You can easily find on the network the real names of the disk models that vendors sell under their brands. For example, most HP SAS drives are made from the Seagate Savvio hard drive line. In this case, it is not at all necessary to search for full analogs; you can choose the appropriate models from the popular lines:


As for the reliability of this or vendor, according to a fairly informative statistics of the company Backblaze , engaged in providing cloud backup, the most reliable are Hitachi disks. Western Digital comes second, Seagate third.

Check warranty


If you decide not to take the lead from vendors and are going to buy "non-native" hard drives, then first be sure to check with the seller: who provides a guarantee? The fact is that many stores do not provide a guarantee for hard drives, referring to the manufacturer's warranty. But there is a delicate point: for example, for some models, the Seagate warranty period starts from the moment of production . Therefore, the situation is not excluded that you buy brand new discs for which the manufacturer’s warranty has already expired.

Sad, but true.

In order not to experience this awkward moment, try to check the warranty of specific copies on the manufacturers websites before buying:

Seagate: http://support.seagate.com/customer/en-US/warranty_validation.jsp

Western Digital: http://support.wdc.com/Warranty/warrantyStatus.aspx?lang=en

Hitachi: https://www.hgst.com/portal/site/en/support/warranty

Conclusion


When choosing a hard drive, you must first build on the tasks that the server will perform:


But the main thing - do not believe the myths. It is not necessary to buy discs with the same logo as on your server. With the right approach, you can significantly save on upgrading the disk subsystem, without losing at all the reliability and speed of work.

Source: https://habr.com/ru/post/302868/

More articles:


All Articles