hi5 saves time with Hyperic Enterprise
I had an opportunity to speak with Akash Garg, hi5 Network’s CTO, about how his organization is using tools from Hyperic. Thanks Akash for taking the time to chat with me.
Please tell me who you are and a bit about your organization.
hi5 networks is a leading social networking destination on the net. They have 35 employees in San Francisco and a service center with additional 10 staff members in India. hi5's goal is helping people connect with friends and to meet new friends.
What does your organization do that needs what Hyperic offers?
At last count hi5 has roughly 30 millions active accounts a month. It offers its service to people in over 200 countries. In any random day, hi5's servers present over 200 million pages. hi5 has over 400 servers to support this large volume of Internet traffic. hi5 supports a large amount of worldwide traffic with a very small customer service staff. A great deal of the ongoing support comes from members who help other members solve usage problems.
What products were considered before Hyperic's product was selected?
hi5 used several open source tools over time. First they used Nagios as a network and system monitoring tool. Later hi5 moved to Big Brother. Both of these tools did a fairly good job of monitoring systems and the network connecting them. They offered quite a bit of information about what's up and what's down. After using these products for a time, hi5 came to the conclusion that they both were really reactionary, eg. they were good for observing if something was up or down. Since hi5 has a very small staff and really doesn't have the resources to spend a great deal of time investigating why failures have happened, they continued to search for a tool that would help them review the historical record of system and network traffic and present this information in a graphical dashboard. With this in mind, hi5 decided to conduct a pilot project with Hyperic Enterprise.
Why was Hyperic's product selected over the competition?
Hyperic gathers and displays quite a bit of historical data making it easier for hi5 to determine what's been happening. They're able to analyze this data to determine the real cause of an issue they're experiencing.
What tangible benefits has your organization gotten from the deployment of Hyperic's product?
On several occasions, hi5 has been able to quickly identify and resolve fast moving issues. The ability to look at graphical views of the data allowed hi5 to rapidly determine what's been happening and indicated where to start the repair process. The number of outages and the time to problem resolution have both been reduced.
Do you have any advice for others who may be facing similar challenges?
Akash pointed out that companies trying to support this type of large network usually end up building their own system and network management tools. He believes that it would be better for them to build their own tools. It would be better to go with a product such as Hyperic Enterprise rather than investing their resources, time and money developing their own custom tools. It's his view that they would save quite a bit of time money.
To the Hyperic community - Does hi5's experience with Hyperic Enterprise match yours? Do you have any other advice to offer to IT decision-makers facing these challenges?