From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roberto Spadim Subject: Re: Mirrored volume peformance questions Date: Tue, 3 May 2011 16:52:07 -0300 Message-ID: References: <96B30EF4F3E17749BCFF7F1816090EDA02DCE41A87@EX-SEA31-B.ant.amazon.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <96B30EF4F3E17749BCFF7F1816090EDA02DCE41A87@EX-SEA31-B.ant.amazon.com> Sender: linux-raid-owner@vger.kernel.org To: "Morad, Steve" Cc: "linux-raid@vger.kernel.org" List-Id: linux-raid.ids 2011/5/3 Morad, Steve : > I have a few questions about volume mirroring performance implication= s. > > 1. I'm looking for an optimal configuration to maximize read speed wh= ile protecting the data from a disk failure (e.g. for an application th= at primarily does reads). =A0For RAID 1, does the linux raid driver per= form parallel read requests across all available disks? =A0If, for exam= ple, I set up a RAID 1 volume with 4 disks, would the OS handle concurr= ently reading (different) data from all 4 disks at the same time? today read_balance algorithm is optimized for multi-thread read algorithm preffer disk with minimal head distance (current block - block that will be read) you can unselect disks setting they as write-mostly optimizations at raid1 aren=B4t for bigger sequencial read, they are fo= r multi-thread > 2. Similarly, would a RAID10 configuration give me the same (or bette= r) read behavior across these same disks, while providing twice the sto= rage capacity of the above configuration? in md world raid1+ raid0 !=3D raid10 raid10 can use layouts raid1 can=B4t raid10 have diferent read_balance algorithms than raid1 raid10 with far layout is better optimized for sequencial read (it=B4s like raid0 stripe) raid10 with near/offset layoute are better optimized for multthread > 3. Is the raid driver smart enough to keep track of underlying volume= performance history or queue lengths and make read requests appropriat= ely? For example, if for some crazy reason a 10K disk was mirrored with= a 7K disk, would the OS make more read requests to the 10K disk than t= o the 7K disk, or would read requests be equally distributed across bot= h disks? no i made a new raid1.c algorithm for kernel 2.6.37, that you can change this options, you will have 1% of speed improvement in a very high workload of read (ok it=B4s not a lot but i tested three times (1 hour non sequencial test) and 1% are seeing, it=B4s not a error of benchmark), you can mix SSD and HD too code is here: http://www.spadim.com.br/raid1/ it=B4s not a md-team solution it=B4s a patch, more information inside the file to compile you must compile like anyother kernel module make drivers/md/raid1 .... > > -- > To unsubscribe from this list: send the line "unsubscribe linux-raid"= in > the body of a message to majordomo@vger.kernel.org > More majordomo info at =A0http://vger.kernel.org/majordomo-info.html > --=20 Roberto Spadim Spadim Technology / SPAEmpresarial -- To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html