Storage Systems for High Scalable Systems Presentation
September 1st, 2009
Storage Systems for High Scalable Systems
The High Scalable Systems (i.e. Websites) such as: Google, Facebook, Amazon, etc. need high scalable storage system that can deal with huge amount of data with high availability and reliability. Building large systems on top of a traditional RDBMS data storage layer is no longer good enough. This presentation explores the landscape of new technologies available today to augment your data layer to improve performance and reliability.
Remember: All of my presentations contents is open source, please feel free to use it, copy it, and re-distribute it as you want.
September 4th, 2009 at 3:29 pm
It’s easy to dismiss the RDBMs as legacy, but look at hadoop and it is doing scheduling, resource management (HOD), locking (Zookeeper), execution plans (pig). We’re just doing it on a different level of scale. The problems: what to run, where are still there.
Also, DBs can scale, just most people run out of money before they top out of oracle’s HDD space. It’s a financial limit, not a technical one. That doesn’t make switching to Hadoop+HBase, Cassandra or Voldemort less valid, because lowering your cost/entry makes saving every log entry forever possible. But most people who say “they’ve reached the limits of a database” really mean “we can’t afford to scale up Oracle or DB2 any more”.
These are the people who come to Hadoop with an unrealistic set of expectations “I want a drop in replacement for Oracle for $0″, have heard the stories, and end up going away disappointed.
September 4th, 2009 at 7:38 pm
[...] Just find a good stuff about scalable storage system, especially Key-Value storage system part. You can download the PDF or PPT format presentation from author blog. [...]
September 5th, 2009 at 4:37 am
[...] Storage Systems for High Scalable Systems Presentation [...]
September 5th, 2009 at 6:45 pm
Nice presentation. Thanks! I shared it on slideshare: http://www.slideshare.net/andyman3000/storage-systems-for-high-scalable-systems-presentation
September 6th, 2009 at 2:44 pm
There’s also Keyspace, which can handle 100.000 ops/sec in 3-way replicated mode, uses Paxos, so it may be used like Chubby in certain environments. We’re using it in production right now.
September 7th, 2009 at 4:16 pm
[...] I expect you to know what is key-Value store system. If you don’t know what is Key-Value store systems I suggest you to read the following posts: – Against all the odds – Drop ACID and Think About Data – Storage Systems for High Scalable Systems presentation [...]
September 7th, 2009 at 10:07 pm
This presentaion does little more than make claims without many facts to back them up. RDBM systems still have plenty of uses in scalable systems if applied properly. Claiming that they are unusable because they were invented decades ago is just silly. Do you believe they are unchanged over that time? Unix is similarly aged. Is that also pointless?
Both types of storage systems have good uses. But only one has a growing band-wagon, which this author is clealy riding.
November 21st, 2009 at 7:36 pm
[...] presentación podeis descargarla desde su entrada de blog en dos formatos: PPTX y [...]