I did some research in last few months and based on my experience I started to use RAID10 for both Vmware / XEN Virtualization and database servers. It provides security by mirroring all data on secondary drives while using striping across each set of drives to speed up data transfers. As to the comments about hard drive space being cheap – please share with me where you’re getting cheap 500G SAS or SCSI hds. Redefine very clear your needs. Two hours is a much smaller window of opportunity for disaster than seventy-two. (1 guys 2 cents worth), Each systems I/O should be well known prior to selecting a storage solution…. Not all your information has same storage needs. * RAID 10 = a striped mirror set In the example 2 TB array above, if I chose RAID 5 and had one disk fail, after that every other disk failure gives 100% chance of array loss. @David The article was short, but helpful. With that being said, don’t be a moron; do your backups. Keep block(cluster) size equal to to data size in stripe for no parity computation penalty. This might be true for some controllers, or even for all current controllers, but it can’t be a long-term fundamental problem. NO RAID5! Thank you! The data in a RAID 10 array is both striped and mirrored. Software as a service via the web, sure, all for it! The key factor is that the storage system and the host should be sharing the load as much as possible. Or are you saying i would still have to set up raid in the new motherboard and that it would automatically recognize the existing raid. I am sure the equations will further change when we have SSD put in.. by Matthew Mister, on Sep 26, 2019 1:35:52 PM. RAID 10 with 4 drives (you can have the fifth drive as hot spare) I think you are mistaken regarding performance of a 4 disk RAID 10. Very curious about this. SCSI or FC disks. However, this comes with the downside of limiting your capacity to 50% of the total disk space. …they dont require you to install drivers during windows installation…. One disk died and while the hot spare was rebuilding (which as you can imagine took ages) we had another die also. One advantage to Raid-10 is that if a drive does go down, you won’t notice a performance hit as you would with a Raid-5 while it has to rebuild the array. – A storage solution without a backup/recovery method and a disaster recovery plan (that fits the within the requirements of the environment) is no solution at all. btw please make the monthly check payable to mr expert”. RAID 50 also provides high data transfer speeds. and no performance cot for parity management. Which is all well and good, until you consider this: as drives increase in size, any drive failure will always be accompanied by a read error. WHAT WILL WE DO IF WE ARE USING HOTSPARE FOR BOTH RAID 50 AND RAID 10 THEN SIMULTANEOUSLY DRIVE FAILS FROM BOTH RAID LEVEL THEN THAT MOMENT HOW THE HOT SPARE WILL WORK.??? With recent drives, try to have partition start at 4k, or bigger power of two allignment on a single drive. Raid 10 is a raid1 array of two raid0 arrays, not a raid0 array of two raid1 arrays. It looks that with increasing HDD capacities RAID 5 will be not able to provide data safety…, Very good article: Why RAID 5 stops working in 2009 at blogs.zdnet.com/storage/?p=162&tag=nl.e539. For a balance of redundancy, disk drive usage and performance RAID 5 or RAID 50 are great options. You have completely overlooked the use of hot spare drives. The articall states Raid 1+0 is cheaper, which isn’t necessarily true. You’ll notice that media streaming or database logs (highly sequential) is where RAID 5 and RAID 6 shine, being outperformed only by RAID 0 on those tests. RAID 0 helps to increase performance by striping volume data across multiple disk drives. Additionaly it is more complex, which nets a more complex and possibly more buggy implementation, and less flexilbility with management. Data loss would result upon disk … In case I’m not being clear, RAID 10 is 1+0, not “ten”. you can lose 1 disk on each raid 1 pair, but not both on the same pair. RAID 0 is the implementation of disk striping without parity. If I am reading this correct, then this article says that a raid 1+0 array offer 4x proformance, when you only have 4 drives. TL;DR: unless your data is not all that important to you do not use RAID as a substitute for backup. Awesome awesome post! Besides, if you’re using SATA in the Enterprise, you deserve the high failure rate. They provide redundancy, allow for the largest range of disk usage and give you data protection that you can rely on. of disks was found in a collection of SATA disks All of it, all the time…NO. By using multiple disks (at least 2) at the same time, this offers superior I/O performance. 4TB space RAID 0 (also known as a stripe set or striped volume) splits ("stripes") data evenly across two or more disks, without parity information, redundancy, or fault tolerance.Since RAID 0 provides no fault tolerance or redundancy, the failure of one drive will cause the entire array to fail; as a result of having data striped across all disks, the failure will result in total data loss. With enough cache power on controller, most of the bane of having parity created is nullified. A Redundant Array of Independent Drives (or Disks), also known as Redundant Array of Inexpensive Drives (or Disks) (RAID) is a term for data storage schemes that divide and replicate data among multiple hard drives. Regards, the cost per gig of drives is so cheap today, that i don’t see a reason to use less than raid 10 if you’re combining multiple disks. When I get rich I may consider a hot spare for the Raid 1 mirrored Western Digital SATA Enterprise drives running SME Linux server from http://www.contribs.org . I intend to use 4 X 2TB SATA II disks. RAID 50 also offers high performance, fault tolerance, and high data transfer rates. The statistical math is wonky, and the numbers are a worst case scenario if and ONLY if you max out the size of the array. In your case with 5 disks, you can consider a RAID5 with 4 data disks and one parity (well parity rotates from disk to disk). RAID 6 seems like the best compromise to me, and I hope that modern hardware controllers with large cache will mitigate the performance issues. “RAID 6 is going to be better than RAID 10 is every way: faster (less duplication needed)”. This post covers up SATA vs SCSI / SAS issue nicely. We can put apps in the cloud, but our data is far too important to entrust to some company out to make money off hosting it. Here is a very simple TLDR chart. For FS block/cluster alligned to physical sector, a partition start at 0x800 should be fine in most cases. Any RAID level will not protect you from multiple disk failures. The standard implementation of RAID 5 may not suit every need but the general approach of using parity vice duplication is in principle sound. Less than 50 percent overhead is important too as I can only purchase and install so many drives and I need to have as much storage space as possible. I have been using RAID 5 on multiple servers for many years. Steve is a moron and clearly has never installed an OS on a RAID array if he thinks you can magically bypass drivers just by changing RAID types. Good performance ( as blocks are striped ). Each RAID level offers a unique combination of performance and redundancy. Regardless if you use RAID0,1,5,10 or any combination, if you rely on them instead of a backup you’re going to have a bad time. RAID 10 however will store 12Mb into 2 drives and then duplicate it – duplicating does not change performance, since drives need to be synced together for consistency. Back the data up from one server to the other – eliminates the tape backup problems. The only thing holding back performance now is the controller, which is of course, these days a waste of money, efficiency and another point of failure. A Raid 5 array starts at almost 2k, not cheap in my book…. The upside of this array is an increase in performance, boasting 2x the read rate of a single disks. A block size of 4 sectors provides data for a complete stipe, that is 4 data sectors and one parity. Either you go offsite, use a NAS/external drive or use a second array of internal drives. Data is currently set up to autocreate multiple volumes for writing to BlueRay so at year end I can archive a nice set of BlueRay disks. RAID 1+0 at least allows for up to 2 drive failures (of course they can’t be part of the same RAID 1). Does a mirrored stripe offer better performance than a striped mirror?? What is impossible/nearly impossible to crack today may not beso hard tomorrow or within a relatively short period of time down the road. You shouldn’t take a crap on something hes adding to the discussion. So 5 disks making 2 logical volumes. If you unlucky and both A1 disks fail, then you have encountered total data lost where with RAID 5 you would need for 3 disks to fail. N/A: Big: Spanning or Concatenation: Data is written on one drive until it is full, and then the next drive(s) until it or they are full. Use this for DB that is heavily read oriented. Sorry folks, data needs to stay local and under YOUR control. please let me know. NO RAID5! RAID 0 stripes all the drives in the array together so a RAID 0’s read and write speeds will be nearly as fast as the combined speed of all the drives in it. My issue is I am building a new high end home personal system and only have experience with raid 0, works great until a drive dies. See here: Which setup do you recommend RAID 01 or RAID 10 ?? Your maximum stripe size is also dependent on the number of disk drives in the array. all represent three-dimensional surfaces of compromise in a discrete four-dimensional space of safety, speed, space and budget. I wonder if anyone has any actual data on failure rates of RAID10. So now what? No one has EVER said that 10 or 50 are implying “better” just because they are a higher number. So many authors of articles I have read from several websites contradict each other and it leaves me wondering who to believe. Maybe I’m right not to believe that there are _IT professionals_ being so dogmatic about such things. This means the performance will be typically worse (although it’s not theoretically much worse, since the parity operations are in parallel). I think it is important to have the terms explained, for someone like myself just getting into RAID configurations on servers. Thanks for the information Kirk, Since I’m not a “tecq” I greatly appreciate the information that you were kind enough to share and the clarity that you helped provide. Take your pick. i’d also like to point out that the linked article about sata references 150’s and not 3.0’s/sata ii specs. next year, I’m going to build a server Highly recommended to those who have 4 or more hard drives in an enclosure (though it can be built with 3x). If you need more than 6 drives in the array for better random access, you will need to use a hardware controller, but these days, people use RAID 6 for large arrays, or even better a RAID aware file-system like ZFS or BTRFS. Scott Similarly, when we talk about raid 60, which is known as raid 6+0 provides you a robust performance boost with a minimum of six hard drives working combined. Reconfigured to raid 0 all 4 disks are happy an running smoothly. That’s why i highly prefer raid 10 over raid 5…. For example: If a drive costs $1000US (and most are far less expensive than that) then switching from a 4 pair RAID10 array to a 5 drive RAID5 array will save 3 drives or $3000US. You can minimize the price difference and play it off, but the difference between RAID 5 and RAID 10 cost as storage space goes up gets pretty hairy. RAID 10 is certainly worth it depending on the context and performance of your data. Does the stress on a drive caused by rebuilding a mirror set make it more likely for a 2nd disk to fail? 1 disk per RAID 1. NO RAID5! Bad IT consultants like you is what drives our business. I think some sort of reliable RAID should be used by all people. You can deploy both implementations separately on the same server and jointly. RAID 5 vs. NO RAID5! He makes a valid point, important for the plebes and c-levels, and accountants who read this. RAID 1 provides disk mirroring which duplicates your data. 4TB space In my case, I had to decide between a 4-disk RAID5 vs RAID10 with 2TB per disk for a home NAS. Not sure if I understand the question but here goes. This performance can be enhanced further by using multiple controllers, ideally one controller per disk. Check out some of our most popular blogs or click below to check out our YouTube Channel. If a RAID10 has a failed drive and another drive hits a URE in the matching mirrored drive reconstruction is going to fail as well. There different types of RAID levels. A solution for all needs is likely to be not the best you can get. Having multiple disks allows the employment of various techniques like disk striping, disk mirroring, and parity.. In this day and age we have to assume any discussion of RAID redundancy and cost effectiveness is centered around SO/HO use, which is where RAID 5 shines brightest. RAID 1+0 is the future but the main point is that RAID is not a replacement for backups. that is part of the LAcie2big and the WD MyBook Studio II products? The major limiting factor in data security is budget. RAID 5 will always give you better price/performance/disk space than RAID 10 for 90-100% READ profiles. “With a 7 drive RAID 5 disk failure, you’ll have 6 remaining 2 TB drives. But are you saying if i have a raid 10 and my motherboard fails i can simply install a new mother board and i’m good to go. The different approaches, etc. I’ve never used any Jetstor products but their I’ve referenced their RAID.EDU website numerous times to students who want to learn more about the different RAID solutions. The upside of this is that parity data is error-correcting redundancy that is designed to re-create data if a drive fails. PS I am not happy, chances of disk failure is now 4 times greater. s = size of individual disk, RAID 5 cost: C((S/s) + 1) I too can’t believe the negativity towards RAID 5 on here. Mirror sets are wonderful when you have a hardware failure that DOESN’T cause some type of corruption in the data structure – however I have more often than not seen that the result is that both mirrors end up with problems, as well as one of the mirrors simply being dead. In general, I’ve found that you can get the best performance for your money if you use RAID 10 on bigger slightly slower disks than you can using RAID 5 on smaller faster disks. Please contact the developer of this form processor to improve this message. In some cases, RAID 10 offers faster data reads and writes than RAID 5 because it does not need to manage parity. So play the odds: when a disk fails, you’re looking at 100% chance total loss if another fails, or 4% chance total loss if another fails? with SSD RAID 1 will be all we need and performance from 1+0 will not be necessary in my opinion. For more details, please feel free to comment and post on my article, When RAID 10 Is Worth The Economic Cost Link. The Cloud as a data storage solution is a bad idea for any data that you would normally keep secured in your own environment. What RAID configuration would you recommend for someone who is doing 3d rendering (ala Maya or 3ds Max) and only has 3 drives to work with? I bill more than that per hour. I like RAID -10 joking I like RAID 0 + RAID 1 simply far more effective than all of the RAID combined. Raid 10 arrays consist of two or more equally sized RAID 1 arrays. Read Christian, and Expert, and read about ZFS. Minimum 3 disks. RAID 5 is not “less than” RAID 10, which incorrectly implies that the higher number is better, instead of demonstrating that it’s a combination RAID set. Oh, you got the opinion of some person in the field, that doesn’t prove anything. RAID 1 vs. Good redundancy ( distributed parity ). I used RAID 5 since the mid-90’s and had no problems. The upsides of this setup is that you are extra protected. E. RAID Level 10. It is important to note that the cache available on the array plays a crucial role in the performance outcome. If your data is not important or you simply don´t care. RAID 0 arrays include two or more disk drives and provide data striping. At least in theory. If you’re going with RAID 0, you better back up your media on another drive because it does not offer data redundan… And when that happens, you are one unhappy camper. Very much the business (big or small) choice when it comes to setting up a storage array. Adding a fourth HDD only cost $100. This blog contains some media management best practices as well as ProMAX Platform Tips & Tricks. This distinction is important for two reasons: 1) RAID 01 isn’t RAID 1, which is what we’d get pronouncing it as a number, and 2) if we imply that higher is better, RAID 50 should be even better still…when it’s just a different way of doing things. It’s a sort of marketing term launched by some RAID vendors, and thus there are differences in implementation from one RAID 6 implementation to another. i agree that sas is a better option, albeit more expensive, but that said; i run multiple low-mid range dell servers on sata 500gig raid 1’s and they’re perfectly fine. Not to change the subject but I was wondering, When I set up raid 10 or if you like 1+0 better ( LOL ) Disk 4 went critical after about an hour of run time. This RAID level provides the highest read data transaction rate and low ratio of ECC (Parity) disks to data disks. There are two types of performance to look at with all storage: reading and writing. Have had several hard drives fail but never a failure of the RAID 5. During the rebuild, the computer was unusable – think defragging your HDD will performing a full AV scan. This just isn’t what “5” vs “10” vs “50” means. Most people as soon as striping and mirroring gets mentioned, you can see their eyes glaze over. While one disk is offline for any reason, your disk array is not fully redundant. http://www.cyberciti.biz/faq/what-happens-when-hard-disk-fails-in-raid-5/. I do use SCSI or SAS for all enterprise servers for speed with high-RPM spindles and cache. RAID LEVEL 5. Second, RAID 5 is all the redundancy any small scale systems administrator will ever need. As soon as one of the drives fails, you will lose all the accumulated information. You cannot compare a 3-drive raid 5 to an 8 drive raid 1+0. In fact it may save you money especially when you consider the performance degradation associated with RAID 5 and high random read / write IO databases such as Exchange. That’s a pretty hefty price increase, especially when you’re on a budget that still has to pay for CPU’s/RAM/etc…. This configuration stripes stored data and parity across all disk drives on both RAID 5 arrays. hit. In our data sets, the replacement rates of SATA I do appreciate the explanation and distinction whereas some others did not. factors, such as operating conditions, I think you are assuming the speed of RAID1 is only 1x for reads, but either member of the mirrored pair can furnish the data, so this is not true; each drive only has to supply half the requested file. RAID 10 is certainly worth it depending on the context and performance of your data. Raid 5? The Cloud is just like the internet in that once it gets placed there you have no way to ensure you’re control over it. These days parity is constructed in NVRAM before being written to disk, its written at the SAME time as any other I/O. We always recommend utilizing RAID in conjunction with an offsite backup package for the best redundancy in your dedicated server. If it is a 4 drive raid 1+0, then it only offers a 2x read speed increase, and this article is incorrect. Eg. RAID 60 arrays provide high data transfer speeds as well. They are linked together to prevent data loss and/or speed up performance. For redundancy this array uses data striping and parity which also provides data protection and a performance boost. RAID-5 1 TB x 4 disks = 3 TB 3X 3X. Please go away and annoy the thicko brigade somewhere else. So the read fails. But it increases t… There is no redundancy in RAID 0; all stored information is evenly distributed across the two disks. RAID 10 cost: C((S/s) * 2) I maintain that the drives are FAR cheaper! The obvious answer, and the one that storage marketers have begun trumpeting, is RAID 6, which protects your data against 2 failures. Some common RAID levels include RAID 0, RAID 1, RAID 5, RAID 10. ZFS maintains a checksum for each byte, and if necessary relocates data on a bad sector to a good sector. The actual performance, though, depends on the controller, and I don’t know enough about the choices there. The performance for a read would indeed be up to 4x. Not all RAID configurations are created equal in terms of redundancy, speed, or disk size. As the RAID controller is busily reading through those 6 disks to reconstruct the data from the failed drive, it is almost certain it will see an URE. RAID 5 costs more for write-intensive applications than RAID 1. It’s widely used for high performance and data redundancy. These units can help you make data redundancy and performance improvement realistic. A parity calculation doesn’t require many operations and just shouldn’t take very long, compared with hard drive write time. I’m sorry, you compared a 3 disk RAID 5 array with a 4 disk RAID 10 array, which is not very helpful. for 4 drives, RAID 5 is in theory, 50% better for both read and write performance. But the performance of this array is the best. Lets agree that RAID-5 is not safe enough on paper, how does it fare in REAL life? So in my opinion it’s 6>5>10>1. Rebuilding a (4 x 2TB RAID 10, on the exact same computeer, took about an hour and a half to two hours (the exact number isn’t burned into my brain, because, while the RAID 10 is rebuilding, except for the first two or three minutes, while the RAID is rebuilding the data at and near track0, performance returns to the usual blazzingly fast RAID 10 performance. ZFS has a ZRAID function which is supposedly better than RAID5. I’m not sure of the theoretical differences of RAID 6 vs RAID 5 with hot standby. Another consideration is to look at host based striping at the same time. Reconstruction next: of course reconstruction will take longer with RAID-DP. Another consideration is to look at host based striping at the same time. My company has done well over 100, probably closer to 200 Raid 5 implementations, at least another 30ish Raid 1’s. Brilliant! RAID level 0 has the best write performance of all RAID levels, because absence of redundant information implies that no redundant information needs to be updated!