Results 1 to 9 of 9
  1. #1

    Upgrading from VPS to VPS SSD? Advice please.

    Hi all,

    Hope all is well.

    We are having some significant speed problems with our website (ttfbaseball.com).

    We are on a baseline VPS with WiredTree right now -- and let me say this: I love WiredTree. The site crawling to a slow is a problem on our end, not theirs. We've simply outgrown our current server.

    Anyways ... take a look at this:

    Code:
     >>  MySQLTuner 1.2.0 - Major Hayden <major@mhtx.net>
     >>  Bug reports, feature requests, and downloads at http://mysqltuner.com/
     >>  Run with '--help' for additional options and output filtering
    
    -------- General Statistics --------------------------------------------------
    [--] Skipped version check for MySQLTuner script
    [OK] Currently running supported MySQL version 5.5.30-MariaDB
    [OK] Operating on 32-bit architecture with less than 2GB RAM
    
    -------- Storage Engine Statistics -------------------------------------------
    [--] Status: +Archive -BDB +Federated +InnoDB -ISAM -NDBCluster
    [--] Data in MyISAM tables: 264M (Tables: 371)
    [--] Data in InnoDB tables: 1G (Tables: 198)
    [--] Data in PERFORMANCE_SCHEMA tables: 0B (Tables: 17)
    [--] Data in MEMORY tables: 0B (Tables: 2)
    [!!] Total fragmented tables: 21
    
    -------- Security Recommendations  -------------------------------------------
    [OK] All database users have passwords assigned
    
    -------- Performance Metrics -------------------------------------------------
    [--] Up for: 21h 45m 7s (930K q [11.889 qps], 29K conn, TX: 8B, RX: 114M)
    [--] Reads / Writes: 86% / 14%
    [--] Total buffers: 720.0M global + 3.4M per thread (100 max threads)
    [OK] Maximum possible memory usage: 1.0G (69% of installed RAM)
    [OK] Slow queries: 0% (507/930K)
    [OK] Highest usage of available connections: 20% (20/100)
    [OK] Key buffer size / total MyISAM indexes: 64.0M/106.6M
    [OK] Key buffer hit rate: 100.0% (14M cached / 4K reads)
    [OK] Query cache efficiency: 50.3% (381K cached / 757K selects)
    [!!] Query cache prunes per day: 32992
    [OK] Sorts requiring temporary tables: 0% (23 temp sorts / 36K sorts)
    [!!] Temporary tables created on disk: 40% (22K on disk / 56K total)
    [OK] Thread cache hit rate: 81% (5K created / 29K connections)
    [OK] Table cache hit rate: 103% (688 open / 663 opened)
    [OK] Open file limit used: 10% (864/8K)
    [OK] Table locks acquired immediately: 99% (481K immediate / 482K locks)
    [!!] InnoDB data size / buffer pool: 1.1G/256.0M
    
    -------- Recommendations -----------------------------------------------------
    General recommendations:
        Run OPTIMIZE TABLE to defragment tables for better performance
        MySQL started within last 24 hours - recommendations may be inaccurate
        Enable the slow query log to troubleshoot bad queries
        Temporary table size is already large - reduce result set size
        Reduce your SELECT DISTINCT queries without LIMIT clauses
    Variables to adjust:
        query_cache_size (> 128M)
        innodb_buffer_pool_size (>= 1G)
    As WiredTree support point out to us, our InnoDB is upwards of 1.1G. That VASTLY looms over our current amount of RAM.

    So, rather than simply upgrading our RAM, it is probably prudent on our part to completely upgrade the server. We are thinking about upgrading to either WiredTree's SSD2000 or SSD3000. The options provide 2GB and 3GB of RAM, respectively. Additionally, WiredTree has me understanding that moving to a SSD-based server will also add quite a boost to our site's speed (especially in the WP-Admin area which is timing out on us all the time now when editing posts/uploading images/etc.).

    I am just hoping to receive some advice on what we should do. I don't wanna upgrade to the SSD server just to find out that there is no significant difference (especially after reading the thread just a few below this one about WiredTree's SSD servers).

    Any and all advice is greatly appreciated!

    Edit to add this: We have LiteSpeed installed, we have switched over to MariaDB, we run W3 Total Cache, we use MaxCDN, and memcached is setup on our server.

  2. #2
    Join Date
    Jan 2007
    Posts
    1,025
    If database is the bottlenet, SSD can help.
    I have moved one of my slow site(mysql problem) to ssd disk, the site loads much much fast on ssd.

  3. #3
    Quote Originally Posted by linuxclark View Post
    If database is the bottlenet, SSD can help.
    I have moved one of my slow site(mysql problem) to ssd disk, the site loads much much fast on ssd.
    If I understand correctly, I believe our problem is database driven ... so your experience is good to hear.

  4. #4
    Join Date
    Jan 2007
    Posts
    1,025
    Quote Originally Posted by BradTTFB View Post
    If I understand correctly, I believe our problem is database driven ... so your experience is good to hear.
    For that site, the web page load time from 5+ seconds to less than 1 second. All I did was moving the site from a non-ssd provider to digitalocean(ssd disk), all the code, database are the same.

  5. #5
    Join Date
    Jan 2010
    Location
    Lithuania
    Posts
    1,115
    Quote Originally Posted by linuxclark View Post
    If database is the bottlenet, SSD can help.
    I have moved one of my slow site(mysql problem) to ssd disk, the site loads much much fast on ssd.
    If you have long and not optimized requests to MySQL database, eventually even SSD wouldn't help.

  6. #6
    Got to be very careful to select SSD VPS provider when you have plan to move on.
    Don't just look at the VPS is "SSD" but have to understand their virt platform. There are many low cost "SSD" host around but got to be understand how many client host on the same. Others, it will just back to square one.
    ControlVM.com :: Elastic Cloud Hosting Alternatives to VPS Reliable Cloud Server
    Enterprise Hosting @
    Malaysia ● Germany ● USA ● Singapore
    We Accept: Paypal
    Alipay GrabPay Credit/Debit Card FPX Bank Transfer Bank TT

  7. #7
    I noticed 5 to 6 seconds waiting time on all your blog pages. Wondering how it is possible when you have W3 Total Cache enabled. W3TC can keep a cached copy of each page on disk and serve from there without needing to regenerate the whole html every time you request. I don't think DB is slowing down your blog pages if pages are served from W3TC cache.

  8. #8
    Quote Originally Posted by mmn357157 View Post
    I noticed 5 to 6 seconds waiting time on all your blog pages. Wondering how it is possible when you have W3 Total Cache enabled. W3TC can keep a cached copy of each page on disk and serve from there without needing to regenerate the whole html every time you request. I don't think DB is slowing down your blog pages if pages are served from W3TC cache.
    Any guesses as to what could be slowing it down if it is not a DB problem?

  9. #9
    Quote Originally Posted by BradTTFB View Post
    Any guesses as to what could be slowing it down if it is not a DB problem?
    Check for any warnings in your W3TC plugin page (from Wordpress admin backend).
    Check your wp-content/w3tc directory and see if page/db caches getting generated.
    Upgrade to latest supported version of W3TC.
    Clear W3TC cache per post/page to see if that helps.
    Check various W3TC settings in plugin page.
    Check wp-content/w3tc/log directory for error logs
    Ask for experts help in WP/other php coders forums.

Similar Threads

  1. Replies: 1
    Last Post: 01-02-2013, 01:28 PM
  2. Upgrading to SSD on VPS
    By mikeymccoy in forum VPS Hosting
    Replies: 5
    Last Post: 08-23-2012, 01:14 AM
  3. Replies: 4
    Last Post: 02-16-2012, 11:46 PM
  4. Replies: 0
    Last Post: 02-10-2012, 01:02 PM
  5. Replies: 4
    Last Post: 02-06-2012, 04:45 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •