How do enterprise hosting plans differ from standard hosting? - Powered by Kayako SupportSuite Help Desk Software

07 Sep 2010

Support Center » Knowledgebase » How do enterprise hosting plans differ from standard hosting?

	How do enterprise hosting plans differ from standard hosting?

Article Our common physical environment is setup as follows for both enterprise & shared hosting customers:

Network peering with:

• PIPE Networks
• Equinix Peering
• Inticon Wholesale
• Ipera Communications

Our primary data centre is the Equinix data centre in Sydney which services most of the Asia Pacific region. Our primary data centre meets the following specifications:

• Minimum N + 1 Power configuration.
• 4 Onsite Generators and a supply of Diesel in case of a full power failure.
• 24 / 7 Security Monitoring, and onsite security staff.
• N + 1 cooling, keeping your equipment at a steady 22 degrees Celsius.
• Biometrics and card access to prevent unauthorised access to the building.
• VESDA (Very Early Smoke Detection Agent) fire system and dry pipe sprinklers in zones, in case of a fire breakout.
• 99.9999 Power SLA
• 99.95 Network SLA

Our network architecture is fully redundant where the network is within our control, i.e. to our edge ISP connections, and is solely based on Cisco networking equipment. We use Cisco’s HSRP to ensure router uptime in the event of hardware failure and use redundant Cisco switches for switch level redundancy.

Our server infrastructure is based on fully redundant HP servers (we do not use whitebox or no-name hardware).

The differences with our enterprise plans from our standard hosting plans are as follows:

1. Dedicated IP addresses for all enterprise plans
2. 24x7 monitoring & reporting of network uptime and utilisation via Cisco Netflow Reporting – Network reporting can be scheduled daily if required, otherwise weekly.
3. Primary and secondary servers setup and data replication setup between servers on an hourly basis.
4. Load balancing provided by a number of different means (selectable):
a. DNS Round robin for HTTP traffic where no session state or preference for servers is required and no intelligence around failure of a web server is expected
b. Where persistent session state is required or intelligence around routing of traffic, we use either software load balancing (no charge) or a highly available hardware load balancer pair where throughput is expected to be significant (custom quote required)
c. A Windows NLB cluster can also be setup, however, is not the preferred solution.
5. Ability to deploy additional servers in other global data centres (custom quote) as a part of the load balanced cluster for geographically relevant serving of content (requires hardware load balancer) or for site level redundancy.
6. Dedicated bandwidth & network links are available to enterprise customers (custom quote)
7. Ability to use burst bandwidth on our network to cope with unexpected traffic
8. Custom infrastructure architecture or design documents can be completed depending on customer requirements but is charged at an hourly rate
9. Phone support provided as per the website (phone support as well as ticket & email – Australian based)
10. SLA’s as per the website.

Article Details

Article ID:

Created On:

23 Sep 2008 7:32 PM