Robin offers a very transformative platform that can be put to many uses. The Robin Containerization Platform has changed the way that some of our users use their infrastructure and also how they deploy their Big Data and other applications on this high-performance infrastructure – Walmart case study. Each group of constituents within a company sees Robin differently and finds surprising new uses for our platform and every time we speak to them a new use case emerges. It would be very difficult to articulate this in one blog post. Hence, I have decided to tackle this through a series of blog posts that will appear in parts over the next few weeks and months.
Today’s blog, the first one in the series, is sort of a summary post that articulates the overall benefits of the platform without going into any specifics. Justice will be done to each topic you see mentioned here, in later posts.
A few words about the Robin architecture before we start elaborating on each of these value propositions. Robin is an application-defined high-performance infrastructure platform that virtualizes bare metal servers using data-aware containers. This infrastructure will let you run any type of Linux application both clustered and non-clustered. Clustered applications typically tend to be Big Data applications such as Hadoop, Cassandra, MongoDB, Spark and Kafka. Non-clustered applications can be microservices such as web servers and processes, databases such as Oracle and MySQL; or your own home-grown application. This level of flexibility is possible because Robin supports both LXC and Docker containers.
Container-based IT Infrastructure
Robin offers an IT high-performance infrastructure platform that is disruptive, re-defines this space, sets standards and helps you realize very tangible CAPEX and OPEX benefits right off the bat, even before you have deployed and run your first application. This has a direct impact on your bottom line.
This is in contrast to offerings that do not change the status quo but keep that firmly in place. Robin’s platform is starkly different from those that bring only incremental benefits and at the same time introduces a host of new operational headaches.
To have a deep impact, the offerings should not only remove all the current pain points related to infrastructure scaling but also mitigate issues related to the applications running on this infrastructure by managing quotas and delivering the level of QoS and SLAs demanded by business and customers.
In other words, you have not solved the problem unless you have addressed the needs of the broadest group of stakeholders within and outside the organization. These include but are not limited to IT Infrastructure teams, business SMEs, IT developments teams, executive, DevOps & support teams and most importantly, your paying customers.
Key Pain Points and benefits for Stakeholders
Listed below are some of the key pain points related to infrastructure and applications and the stakeholders who have a say and will benefit from any type of optimization that the product can bring to bear to help mitigate the issues that follow.
|Team Needs||IT Infrastructure Team|
|Business Users & IT Dev Teams|
|DevOPS & Support|
|Agility: Complete environment, additional provisioning capacity.||✓||✓||✓||✓|
|Multi-tenancy: Ability to run multiple isolated clusters on same shared physical hardware.||✓||✓||✓|
|Data sharing among applications.|
|Single-click provisioning: Spin up multi-node Hadoop clusters with a single command from the command line or GUI.|
|Caching: Workload benefit from compute-side SSD caching.|
|Elasticity with separation of Compute and Storage where the two layers can grow or shrink, independent of each other. This also ensures that you do not end up paying for storage or compute resources when you do not need them.|
|Lower cost through reduced H/W footprint resulting in reduced data center storage requirements and reduced power consumption. Reduced hardware footprint also means less number of software licenses required.|
|Reduced hardware footprint and energy costs with multi-tenancy and co-location.|
|Enterprise-class high availability – tolerate disk, node, rack and application failures. Ensure business continuity.|
|High level of Flexibility and Elasticity|
- Hybrid architecture with compute and storage on the same node
- Heterogeneous hardware configurations with core dense machines for CPU intensive jobs
- Ability to add virtual instances to existing clusters on the fly
- Ability to add physical nodes and spin them up with application components on the fly
|Access to telemetry information to perform diagnostic and predictive analysis.|
|Versatile GUI – single-click provisioning, dashboards.|
|Very effective monitoring system with alerts for quick detection of anomalies, self-healing and quick alerts for any required manual intervention.|
So far if you have been thinking that I am preaching to the choir and indulging in some form of pontification, allow me to explain and elaborate.
All of the aforementioned features and capabilities are available out of the box with Robin. A lot of these were also some of the key pain points and asks from one of our major retail clients, which is in the process of deploying Robin to production. This includes a lot of their bread and butter applications which will use Robin as their primary operating platform. Their existing infrastructure will now act as their DR platform.
In my next blog post, I will describe several customer use cases wherein they built a very agile platform with a focus on the agility aspect of Robin.