Mesos is an open-source cluster manager that allows for resource isolation and sharing across distributed applications, or frameworks. The integration with Mesos automates the data center infrastructure beneath Riak KV instances. Pairing Mesos with Riak's own automation and orchestration technology enables enterprises to deploy distributed data services at global scale in support of Internet of things (IoT) and big data applications, while ensuring efficient utilization of cloud resources.
"Companies using traditional architectures are struggling to accommodate the need for distributed applications and distributed data services," said Dave McCrory, CTO at Basho, in a statement. "By combining Basho's Riak KV with Mesos, we're able to deliver an easy-to-deploy platform for real-time data processing. We thereby enable a new class of modern data center developers who can break free of infrastructure restraints and give rise to a whole new class of hyper-scale applications."
Riak KV is a highly available, distributed database. By integrating with Mesos, users gain the benefit of Mesos' resource management, so they no longer have to guess or estimate about the infrastructure requirements of the Riak nodes. With Riak KV managing the data tier and Mesos managing the underlying infrastructure, enterprises now have access to an efficient and easily scalable distributed data platform. The integration also allows for true "push button" scale up/scale down as Mesos can aggregate and re-aggregate resources for or from Riak nodes. Users will be able to write scripts to make scale up/down events automatic based on business logic, enabling enterprises to auto-scale a global, multi-data center database.
"Enabling Riak KV with Mesos on Intercloud, we can seamlessly and efficiently manage the cloud resources required by a globally scalable NoSQL database, allowing us to provide the back-end for large-scale data processing, web, mobile and IoT applications," said Ken Owens, CTO for Cisco Intercloud Services, in a statement. "This integration will accelerate developers' ability to create innovative new cloud services for the Intercloud — the globally connected network of clouds Cisco is building with its partners, which will offer cloud services to help customers capture the multitude of opportunities created by the Internet of Everything (IoE). We're making it easier for customers to develop and deploy highly complex, distributed applications for big data and IoT."
Mesos, which provides efficient resource scheduling and sharing across multi-data center environments, decides how many resources to offer each framework, like Riak KV, and what application to execute on the available physical resources. With Mesos integration, users can easily scale an application from one to thousands of instances in minutes. Basho is developing an open-source integration with Mesos that will also be commercialized around a supported enterprise offering. Basho also plans to incorporate the resource management capabilities of Mesos as a core service offering of the Basho Data Platform.
"Our customers implement Riak KV for its characteristics of scalability and fault-tolerance," said Tyler Hannan, director of technical marketing at Basho, in a blog post. "If Mesos can be used to assist in the scalability, that is a positive outcome. But it cannot come at the expense of fault-tolerance. To that end, the Riak Mesos Framework scheduler, at present, attempts to spread Riak nodes across as many different Mesos agents as possible to increase fault tolerance. If there are more nodes requested than there are agents available, the scheduler will then start adding more Riak nodes to existing agents."
Basho demonstrated the beta framework for Riak KV running on Mesos at MesosCon last week in Seattle.
In May, Basho introduced the Basho Data Platform, which integrates Riak KV and Riak S2 (formerly Riak CS) object storage software with Apache Spark for in-memory analytics, Redis caching for increased performance and Apache Solr for enhanced query capabilities, significantly enhancing productivity for development and operations teams.
"Big data with multiple data models, hybrid cloud architectures with distributed active workloads and the rise of the Internet of Things require developers to integrate, replicate and synchronize information across functions," said Mac Devine, vice president and CTO of IBM Cloud Services, in a statement. "Distributed systems are at the convergence of big data, cloud and the Internet of Things. Basho specializes in solving distributed systems challenges, and integrated approaches such as the Basho Data Platform help ensure that applications are highly available, massively scalable and easy to deploy at production scale."