Apache Mesos is a rising name in the world of cloud computing and datacenter deployment and management.
Apache Mesos is a cluster manager which provides efficient resource isolation and sharing across distributed applications or frameworks. It has some amazing features which makes managing your datacenter a breeze. Apache Mesos is completely open source and free, and it was originally developed at the University of California at Berkeley. As great as Apache Mesos is - it is not the most user-friendly or easy software to use.
The arrival of Mesosphere DC/OS has changed all that. Mesosphere DC/OS is a very user-friendly wrapper based on Apache Mesos. It makes it super easy to deploy and manage your Apache Mesos cluster. DC/OS is an enterprise grade datacenter-scale operating system which provides a single, robust and scalable platform for running containers, big data, and distributed apps in production.
Managing resources at the massive scales on which datacentres work has always been a challenge, adding unwanted complication. Apache Mesos and DC/OS are making it drastically easy to manage resources efficiently and scale elastically, which is why organizations such as Verizon, Twitter, and Microsoft chose them to power their datacentres and cloud services.
So how does Mesosphere go about making life easier for its users? As mentioned above - at the base of DC/OS is the open source Apache Mesos engine and a smarter, automated way of handling applications and data. The Mesos engine itself has been tested in real world scenarios through implementations by big names such as Twitter, Apple, and Yelp. DC/OS has native support for new and more efficient packaging methods like Docker containers which can be used instead of traditional virtual machines using a fraction of the resources. This also makes the system ideal for modern apps that work on a distributed infrastructure, utilizing a variety of services ranging from data storage to data analysis.
The other major advantage of Mesosphere DC/OS is the two layer scheduler architecture that makes it far more suitable for scaling and introducing new services. Traditional schedulers have a hard time coping with maintaining an application’s state as well the state of the underlying infrastructure while also optimizing the workload. With the Mesos engine’s dual architecture, these two tasks are divided up and performed separately far more efficiently. The result is that Mesos can easily handle a variety of workloads ranging from container orchestration to big data technologies and even operating systems, drastically improving utilization. Moreover, with a dedicated layer handling the infrastructure logic, scaling is far easier making it extremely well suited for modern applications in their deployment phase.
Since both the Apache Mesos engine and Mesosphere DC/OS are open source and freely available, startups, companies, and users can easily get started with this stack without shelling out a huge software licensing fee. Moreover, since these projects are open source - you do have access to the source code and you are free to change the code, contribute to the community and optimize the frameworks even further.
The consequence of all these is a fully orchestrated and automated network of servers and a more efficient way of managing the applications and associated infrastructure.
Verizon, one of the largest telecom service providers, is one of the largest companies to take advantage of the freedom that Mesosphere offers. Due to the high service availability requirements for telecom networks, their datacentres are over compensated and therefore run at quite low utilization. The use of containers and managing their allotment automatically using Mesosphere, Verizon can allow applications to run more efficiently and save potentially millions in hardware costs alone. Moreover, the elasticity built in allows it to scale automatically to meet peak loads without any degradation in services.
Twitter was also one of the showcases of this technology. While trying to scale up, their engineers turned to Mesos to do it more efficiently. However, with modern applications and the use of a myriad of distributed services, this is a quite complicated task. Mesos allowed the engineers at Twitter to use an entire datacentre with all its resources much like your average desktop makes it far simpler to run applications on them efficiently.
Microsoft too has recently shown interest in the capabilities offered by Mesosphere by partnering with it to run the container based service on Microsoft Azure. For the cloud services provider, efficient use of datacentre resources is the name of the game.
Mesosphere allows safe and reliable operation for services while utilizing infrastructure more efficiently. Add in the ability to scale and the open source nature, and you have an all in one solution that makes Mesosphere an optimal choice for the cloud services giant.
So in essence, Mesos works by simplifying the way that datacentres are deployed, managed and scaled. Without the unnecessary overhead of virtualization, it leverages the power of container technology. Moreover, by the use of automated schedulers to manage applications and infrastructure separately, Mesosphere provides a far more efficient alternate to traditional datacentre management systems.