Performance and Trending

Learn more…

The most disruptive technology is an educated consumer.

Check out our White Papers to learn about OaaS, how to be up 100% of the time (INCLUDING MAINTANENCE), increase profitability, scale on demand, etc.

Read our White Papers >

Why does Performance trending matter?

What good would a car be if it didn’t have an instrument panel?  How effectively could you operate a vehicle with no speedometer or gas gauge, engine warning light, or temp sensor?  How about an airplane?  Imagine trying to fly a plane with no artificial horizon, GPS, or altimeter.

This is exactly what cloud solution providers are delivering when they deploy cloud instances with no SNMP based performance trending analysis.  Cloud providers deliberately obfuscate what is going on “under the hood” to increase the amount of over-subscription potential for each server and to artificially pump revenue by leaving the only choice for customers when performance is slow, to “buy more.”

At Peak Hosting, we feel the knowledge is power, and we empower our customers with the full details of every aspect of their system’s performance so they know:

  1. Where the hot spots are in the hardware
  2. What is the limiting performance factor
  3. Where we should be tuning or changing the architecture for the best return on investment
  4. What the affects or application changes are on their infrastructure.

None of this is possible without detailed SNMP-based performance trending statistics on the hardware’s disk, CPU, memory, and network. However, this is only the start.  Every company, to successfully scale and understand their applications, needs to also have detailed performance trending of their Operating System and individual applications running.  Without knowing the details of how the OS and applications are scaling, it’s impossible to understand what the proper scaling choices are.  “Buy More” works great for the bottom line of the hosting provider, but is very seldom the right answer.

What if I need custom monitoring?

No problem!  If customers need graphing for custom SNMP OID (Simple Network Management Protocol Object Identifier), they only need to reach out to our 24 x 7 support staff and we will gladly add this monitoring to one or all of their servers.

Here is a partial list of just *some* of the SNMP performance trending data Peak Hosting provides:

 

  • apache
    • Apache processes
  • disk
    • Disk IOs per device
    • Disk latency per device
    • Disk throughput per device
    • Disk usage in percent
    • Disk utilization per device
    • Inode usage in percent
    • IO Service time
    • IOstat
  • network
    • Firewall Throughput
    • HTTP loadtime of a page
    • Netstat
  • postfix
    • Postfix bytes throughput
    • Postfix Mailqueue
  • processes
    • Fork rate
    • Number of threads
    • Processes
    • Processes priority
    • VMstat
  • system
    • Available entropy
    • CPU usage
    • File table usage
    • Individual interrupts
    • Inode table usage
    • Interrupts and context switches
    • Load average
    • Logged in users
    • Memory usage
    • Swap in/out
    • Uptime
  • time
    • NTP kernel PLL estimated error (secs)
    • NTP kernel PLL frequency (ppm + 0)
    • NTP kernel PLL offset (secs)
    • NTP timing statistics for system peer
  • diskstats_iops
    • disk
      • Disk IOs for /dev/sda
  • diskstats_latency
    • disk
      • Disk latency for /dev/sda
  • diskstats_throughput
    • disk
      • Disk throughput for /dev/sda
  • diskstats_utilization
    • disk
      • Disk utilization for /dev/sda