Score:0

Is a blade server a cluster system or a multiprocessor system?

mx flag

I'm reading the book "Operating Systems Concepts" by Silbershatz et al. and I'm a bit confused about blade servers, in section 1.3.2, blade servers are classified under "multiprocessor systems" but the definition of a blade server is a chassis containing Multiple boards, each with its own operating system and this operating system may be multiprocessor. But then, a clustered system is defined as individual systems joined together, doesn't that sound like what a blade server is?

The book does acknowledge that there is a debate over the concrete definition of a clustered system, so are these servers considered as both types of systems?

Score:3
ar flag

A blade server is simply a chassis with individual blades that share some common elements, such as the chassis, power supply, cooling, and related cabling.

Each blade is its own server, with it's own CPU and RAM, and possibly storage. Storage may of course be provided by a SAN solution over e.g. iSCSI.

The blades operate as individual servers. Each blade may or may not have multiple CPU sockets, but today all modern CPU's are multicore, so they will effectively be multi-CPU-systems. Blades in one or more chassis may be configured as a cluster, if it's convenient for the workload.

Computer racks are commonly 19", and the minimum equipment size is 1U. For a lot of simple servers, this is simply overkill. By placing them into a chassis that is mounted in a 19" rack, density can be increased quite a lot.

Abdul Munem avatar
mx flag
Oh that clears up my confusion about the definition, thank you! Also just a quick question (sorry if this is silly but I'm an undergrad so I'm not familiar with this) but does the chassis do anything else other than provide a means to connect the blades to a SAN or a power source etc?
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.