Oracle RAC - SQL Server equivalent??

  • Anybody used Oracle RAC ?? Are there any benefits - Is there similar options in SQL Server ?

    Cheers guys

    --------------------------------------------------------------------------------

    Justin Hostettler-Davies

    http://www.databaseexpertise.com

    [font="Verdana"]Justin Hostettler-Davies[/font]
    www.databaseexpertise.com

  • There's no scale-out options for a single SQL Server database. If you have a datawarehouse (star/snowflake schema) check out "SQL Server 2008 R2 Parallel datawarehouse edition" formerly known as Project Madison.

    Gail Shaw
    Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
    SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

    We walk in the dark places no others will enter
    We stand on the bridge and no one may pass
  • My understanding of Oracle RAC is that is is a 'Shared data' scaleout. There is nothing equivalent in SQL Server, the closest is the 'shared nothing' scaleout offered by SQL Server 2008 R2 Parallel Edition. IBM also offer DB2 in shared data and shared nothing scaleouts. All these solutions are designed for high-end workloads.

    There is a lot of stuff written by theorists and people with practical experience of different technologies about the relative benefits and drawbacks of shared database compared to shared nothing. Just about the only thing the two camps agree on is that if you can get your workload to fit on a single machine it will be cheaper to implement and easier to manage than a scaleout solution.

    With the capabilities of today's commodity servers going up to 24 cores and 128GB memory on a single blade, you really only need to worry about scaleout if you are managing over (say) 3TB data - anything smaller and you should be OK with a top-end blade costing about GBP £40K and a single SQL Server license.

    The other reason people use scaleout is to service a web front-end. This is typically done by replicating copies of your master database to multiple servers, each serving the same contents to the web. This aproach normally uses low-cost blades, and is a completely different issue to the problem being tackled by Oracle RAC and SQL 2008 R2 Parallel Edition.

    Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.

    When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara

  • Hi JHD,

    Oracle RAC (Real Application Clusters) corresponds to SQL Server Clustering. The main difference is that RAC allows for multiple active nodes (active/active), especially in more recent releases of Oracle (http://www.oracle.com/technology/products/database/clustering/pdf/twp_rac11gR2.pdf), but SQL Server 2008 Clustering is active/passive only (http://msdn.microsoft.com/en-us/library/ms189134.aspx). RAC also requires additional licensing, which SQL Server doesn't. HTH

  • Nice Marmot (2/8/2010)


    Hi JHD,

    Oracle RAC (Real Application Clusters) corresponds to SQL Server Clustering.

    No it doesn't. Windows clusters are failover-only for resilience and availability, whereas RAC is a MPP technology for improved scalability. So they address a totally different set of requirements.

    The SQL Server equivalent would be to use a middle tier layer as a load balancer between a farm of servers. Or for decision-support applications you could use 2008R2 PDW:

    http://www.microsoft.com/sqlserver/2008/en/us/parallel-data-warehouse.aspx

  • Nice Marmot (2/8/2010)


    Oracle RAC (Real Application Clusters) corresponds to SQL Server Clustering

    This is incorrect. As far as I know there is nothing in the SQL Server side providing Oracle RAC alike functionality.

    Oracle RAC provides both workload balancing and high availability.

    In short, the basic architecture is two to four nodes -Oracle instances- pointing to a single, shared storage area. Nodes talk to each other mostly for data buffering, locking and crash recovery purposes.

    Hope this helps.

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • With the capabilities of today's commodity servers going up to 24 cores and 128GB memory on a single blade, you really only need to worry about scaleout if you are managing over (say) 3TB data - anything smaller and you should be OK with a top-end blade costing about GBP £40K and a single SQL Server license.

    vending locators[/url]

    vending locator[/url]

    bulk vending locators[/url]

  • amrakhjaaa (4/23/2010)


    With the capabilities of today's commodity servers going up to 24 cores and 128GB memory on a single blade, you really only need to worry about scaleout if you are managing over (say) 3TB data - anything smaller and you should be OK with a top-end blade costing about GBP £40K and a single SQL Server license.

    Availability, that's the key word.

    With a single computer you have a single point of failure - Oracle RAC provides high availability by having several independent - yet coordinated- nodes pointing to the same database. One node was down, remaining nodes take the workload - and rollback transactions active on dummed node at the point of failure.

    Looking further, it is not just high availability. RAC also allows you to shutdown one node at a time to conduct maintenance at host level; you can maintain your hardware/software with zero database downtime.

    Hope this clarifies.

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • The other reason people use scaleout is to service a web front-end.

    wedding pearls - website search optimization

  • anna_lyn00 (4/30/2010)


    The other reason people use scaleout is to service a web front-end.

    Being this a SQL Server forum not a web services one this has to be spam, isn't it?

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • The closest thing in SQL would be geo-clustering which offers multi-datacentre redendancy in a combined HA/DR solution. But it certainly does not offer load balancing. And geo-clustering is not as easy to set up as a solution as Oracle RAC.

    Perhaps the next full edition of SQL will offer a form of load balancing/

    /* ----------------------------- */
    Tochter aus Elysium, Wir betreten feuertrunken, Himmlische, dein Heiligtum!

  • GregoryF (4/30/2010)


    The closest thing in SQL would be geo-clustering which offers multi-datacentre redendancy in a combined HA/DR solution. But it certainly does not offer load balancing. And geo-clustering is not as easy to set up as a solution as Oracle RAC.

    If allowed, I would say this is closer to Oracle's Data Guard - failover solution for DR, than RAC.

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • This was removed by the editor as SPAM

  • This was removed by the editor as SPAM

Viewing 14 posts - 1 through 13 (of 13 total)

You must be logged in to reply to this topic. Login to reply