In my last blog post I focused on our vision for the year 2020, and what are the core sets of challenges that IT is facing in 2015. The primary challenge is the proliferation of infrastructure silos of different hardware and computing platforms. This leads to highly inefficient resource utilization, despite wide adoption of server virtualization. Server virtualization combined with the emergence of new compute platforms has further caused storage silos.
Springpath Data Platform is a singular data platform that addresses the storage needs of various applications running on top of compute platforms, such as virtualization, containers and physical servers.
This blog post takes a look at the key aspects that make up Springpath Data Platform. There are three major components to the platform:
1) Data Access Layer
2) Application Programming Interface (APIs) for cloud based management
3) HALO (Hardware Agnostic Log-structured Objects)
The Data Access layer provides a common abstraction to be used by compute platforms regardless of file or block. In addition, the platform can expose an Object store and can also provide direct access to Hadoop environments through an HDFS plugin. This broad access support means a wide variety of application environments can be supported.
The APIs enable a cloud based management environment. This allows enterprise customers to manage their IT infrastructure using cloud-based portal. The data platform reports and logs alerts, events and the periodic health of the system to a centralized log collection service in the cloud. This allows customers to maintain, monitor and view their system as well as enables automatic filing of trouble ticket with Springpath support team when an infrastructure has issues.
HALO, is a completely homegrown, distributed file system built from ground up. HALO has been written in user-space (POSIX) to allow complete portability and ease of upgrade. HALO is a multi-layer structure and we will delve into its layers with my next blog post.