Los Angeles hosting company Media Temple is offering customers a one-month credit after its grid hosting service suffered persistent problems with latency earlier this week. The problems began Oct. 1, and required a series of expansions of grid servers and storage 
to resolve. “Our primary goal is for an outage like this to never happen again,” the company said in offering the customer credit.
The outage at Media Temple underscores the need for strong data center capacity planning 
. Media Temple attributed the latency problems to “unexpected growth causing a general lack of computational resources.” Problems continued after the company boosted hardware nodes on the grid by 25 percent, doubled the RAM in every node, and redistributed storage systems.
Customers received regular updates on the company blog, and Media Temple says it has learned lessons that will help it refine its capacity planning. “(mt) Engineers have come up with new ways to measure Grid performance and will be adding them to our monitoring systems over the next few days, increasing the likelihood that we will detect symptoms before they affect customers,” the company wrote. “We have also changed our growth projection formulas so they will better predict when we need to add hardware to the clusters, avoiding issues like this in the future.”
Media Temple is among the hosting companies that have launched grid services to offer scalable hosting to customers. Others include Mosso, Layered Technologies, ServePath and Concentric. Improved reliability has been a major selling point for grid hosting services, including Media Temple, which has described its grid service as “designed to withstand numerous hardware and software failures without impacting any aspect of your online operations.”
Article printed from Data Center Knowledge: http://www.datacenterknowledge.com
URL to article: http://www.datacenterknowledge.com/archives/2007/10/05/lag-problems-for-media-temple-grid/
URLs in this post:
 series of expansions of grid servers and storage: http://weblog.mediatemple.net/weblog/category/system-incidents/web-and-email-latency-on-gs02-grid-service-cluster-2/
 data center capacity planning: http://www.datacenterknowledge.com/archives/2007/Sep/19/data_center_capacity_management_heats_up.html
 Rich Miller: http://www.datacenterknowledge.com/archives/author/richm/
Click here to print.