Bigstep Launches High-Performance, Low-Latency Spark-as-a-Service for Real-Time Streaming Applications

Bigstep, the big data cloud provider, today launched a bare-metal Spark-as-a-Service offering. The Bigstep Real-Time Spark Service provides a high-performance, low-latency and scalable environment for the rapid design and deployment of real-time data-streaming applications.

Big data research firm Wikibon says Spark-based investments will grow from six percent of total big data spending in 2016 to 37 percent by 2022, and that half of all Spark implementations are already in the cloud. Unique among these cloud solutions for Spark, Bigstep’s is purpose-built for the high performance demands of fast-moving and large-scale real-time streaming projects and requires a minimal IT footprint.

Apache Spark has emerged as the technology of choice as more and more companies build real-time streaming solutions for customer targeting, IOT applications or algorithmic decision-making,” said Bigstep Founder and CEO Lucas Roh. “But for all its power, many IT organizations struggle to build and scale Spark-based infrastructure effectively, whether for ad-hoc work by data scientists or high-performance business applications. Because it is built on our bare-metal cloud platform, the Bigstep Real-Time Spark Service greatly reduces development time, maintenance effort and operating costs. It is easily deployed, infinitely scalable and high-performance.”

Bigstep’s offering is a pay-per-use, fully managed, auto-scaled container-based Spark cluster, optimized for real-time streaming applications that use multiple concurrent Spark contexts. It offers the following benefits:

• Low Latency: Runs on low-latency 40Gbps software-defined bare-metal network fabric.
• Auto-Scaling: Instantly scales Spark clusters and underlying physical infrastructure.
• High-Performance: Single-tenant, hypervisor-free servers for exceptional performance.
• Simple Management: Control center makes it easy to run unlimited Spark clusters simultaneously.
• Rapid Prototyping: Features built-in Jupyter interface for Scala, Python and R.
• Multi-Version Compatible: Runs multiple versions, including Spark 2.0, on same pool of resources.
• Application Integration: Runs side-by-side with other containerized application clusters.

Sign up for the free insideBIGDATA newsletter.