This is a continuation of a series of articles on hosting solutions and services on Azure public cloud with the most recent discussion on Multitenancy here. The previous articles introduced multitenancy via hyperconvergence specifically with examples of on-premises technologies. This article continues with a specific example of virtual SAN.

Storage industry and databases have a rich history and tradition of providing multitenancy. As the hardware and software evolved and transformed the storage industry, hyperconvergence and multitenancy also changed. While there are many examples and products that have their own tales to tell as the exhibits in a museum, this article studies virtual SAN and the databases it powered.

Traditional databases particularly the eCommerce catalogs of enterprises required large Storage Area Networking (SAN) because the databases were designed to persist everything to disk from the database objects, prepared plans, materialized views and others. When the disks of a few gigabytes don’t suffice, Storage Area Networking offered the possibility of near limitless storage for that perspective. vSAN took it a notch higher with its ability to virtualize SAN over several devices.

A market leader in vSAN, for instance, pools together server-attached storage (SSDs, HDDs and other flash devices). It creates a shared data store with advanced data services designed for virtual environments. This datastore is highly resilient with no single point of failure. It is optimized for the latest flash technologies. Spread over many VMs with these disks, the virtual SAN can expand to large storage capabilities tolerating failures and providing a single shared datastore.

This vSAN provider integrates the virtual SAN by building it into the kernel of the hypervisor. Hyperconverged solutions use the hypervisor to support and deliver storage functions and storage networking in software eliminating the need for dedicated storage hardware. Since it is embedded in the kernel, it can deliver the highest levels of performance without taxing the CPU with additional overhead. The in-kernel architecture also simplifies management and eliminates risk associated with extra components and points of integration. In this way, it differs from the many virtualized storage appliances that run separately on top of the hypervisor.

Since storage is a key contributor to performance and efficiency, the load passed on by the hypervisor to virtual SAN storage must be dealt with adequately. In this regard, the vSAN is matured over the Flash storage for nearly a decade. The software that used to be implemented on a disk array, moves onto the hosts. A hyperconverged storage is built from the grounds up to integrate and leverage all the functionality of the hypervisor without operational overhead or any reduction of core functionality. The virtualization layer provides features such as high availability and the live migration of running virtual machines from one physical server to another with zero downtime, continuous service availability and complete transaction integrity. This creates a dynamic, automated, and self-optimizing data center.

 

#algorithms

Determine cycles in a graph

Bellman-Ford(G, w, s)
Initialize-single-source(G,s)
for i = 1 to number of vertices -1
    for each edge (u,v) belongs to edges E in G
          relax(u,v,w)
for each edge(u,v) belongs to edges E in G
     if (v.d  > u.d  + w(u,v))
         return False
return True

Friends Pairing problem:

Given n friends, each one can remain single or can be paired up with some other friend. Each friend can be paired only once so ordering is irrelevant

The total number of ways in which the friends can be paired is given by;

Int GetPairs(int n)

{

If (n <=2) return n;

Return GetPairs(n-1) + GetPairs(n-2)*(n-1);

}