Industry interest in Wikibon’s definition of the new Server SAN architecture has been strong. Top industry podcasts in cloud (The CloudCast), virtualization (VMware Community Roundtable) and storage (SearchStorage) have had me as a guest to talk about the topic. This week, we held a CrowdChat, co-hosted by Howard Marks, Gunnar Berger and myself, on Server SAN to educate on the topic and answer questions (see below for the embedded results).

Scalable converged compute+storage

There is ample evidence that the trends and definition that Wikibon has put forth are resonating with the community. It is an important topic that spans multiple disciplines and the CrowdChat successfully engaged many people across a large number of topics during the hour. While end-users are often hesitant to adopt new technology, the trends discussed in Server SAN have been percolating for years and the resulting solutions are compelling in that they promise to simplify operations while lowering both CapEx and OpEx.

Server SAN is a systems architecture, so each discipline (storage, networking, virtualization) brings its own perspective that can limit acceptance of a new approach that takes a holistic view of the solution (which is critical for CIOs).

There is broad agreement in the trends and the value of the new architecture. It seems that the most controversial piece of the topic is the term “Server SAN”; which in itself draws more attention to the conversation. The genesis of the name is that Server SAN is a new version of both DAS (leveraging internal storage) and SAN (sharing of storage with a trusted and reliable architecture) by creating a new architecture of software, advanced storage (flash and high capacity disk), compute (dense, multi-core) and high-speed, low-latency interconnect (such as RDMA over Ethernet/RoCE) that was not possible only a few years ago.