From the individual server, all the way through to the network, MediaLayer utilizes nothing but the best infrastructure available.
All of our Application Hosting and Application Intensive servers are connected via dual-gigabit connections (public and private VLAN) eliminating network-side bottlenecks. Some providers are still using connections as low as 10mbps. As residential cable/fiber connections are often even faster than this, this can cause a shortage of bandwidth capacity. With connections as fast as 1000mbps, this possibility is virtually eliminated.
In all servers MediaLayer uses RAID arrays consisting, typically consisting of 4 or more hard disks which not only increases reliability, but performance as well. Our most commonly used RAID level is RAID 10, which functions by striping a pair of two (or more) mirrors, allowing the potential to withstand up to half of the hard disks failing, and also the ability to handler a greater number of I/O (input/outpit) operations per second, drastically increasing disk performance.
While many providers still operate using older or desktop-grade CPUs, MediaLayer uses the latest Intel Xeon processors, including some of the fastest Xeon Nehalem Quad Core chips as well as Hex-Core chips in certain systems. With these advanced chips, instead of simply increasing server density (by placing more customers on a server), we continue to maintain a low contention ratio, allowing all of our customers to enjoy the performance boost the latest technology has to offer.
We currently operate a constantly expanding cluster of DNS servers dispersed throughout the globe. We will soon be implementing IP anycast to further increase performance and reliability and facilitate future CDN services. Access to these DNS servers comes standard with all of our offerings, allowing you to have redundant DNS for your site at no additional cost.
Our Application Intensive offering features the best-in-class cloud server architecture featuring a redundant SAN which uses a caching layer of SSD storage to provide the highest level of performance. In addition, with multiple nodes handling virtual servers we are able to achieve high availability - should a node fail we can immediately restart virtual servers on another node. With the ability to perform seamless migrations between nodes, we can also add additional CPU cores and RAM to any virtual server on-demand to handle additional traffic. This level of infrastructure is available in three continents, featuring major cities such as New York, Amsterdam, and Singapore.