Questions

1. What is Platform Symphony MapReduce?
2. What is the business problem Platform Symphony MapReduce solves?
3. What is unique about Platform Symphony MapReduce?
4. Does Platform Symphony MapReduce run MapReduce jobs faster?
5. Why is supporting multiple applications on a single distributed file system so important?
6. What are the limits of Platform Symphony MapReduce?
7. If a customer does not use Platform Symphony MapReduce, what do they lose?
8. Is Platform Symphony MapReduce compatible with open source Hadoop stack?
9. How does Platform Symphony MapReduce compare to open source-based solutions?

Answers

1. What is Platform Symphony MapReduce?

   Platform Symphony MapReduce is a best-of-breed distributed runtime engine developed by Platform Computing.  performing analytics on unstructured data that is maintained on a distributed file system It is used for managing and scheduling map and reduce jobs produced from applications using the MapReduce API, for. Back to top

2. What is the business problem Platform Symphony MapReduce solves?

   Platform Symphony MapReduce helps to significantly reduce the total cost of ownership for MapReduce applications using a distributed file system, through, a) shared applications workloads (no silo’s) and b) higher resource utilization (SLA.) Back to top

3. What is unique about Platform Symphony MapReduce?

   Unlike alternative open source runtime engines, Platform Symphony MapReduce is unique in three areas: a) meets enterprise IT requirements by delivering high resource utilization, availability, scalability, manageability, and service level agreements, b) has an open architecture to support both commercial and open source MapReduce applications as well as other distributed workloads on a common set of resources, and c) supports multiple distributed file systems, including HDFS, IBM GPFS, Appistry CloudIQ, etc. Back to top

4. Does Platform Symphony MapReduce run MapReduce jobs faster?

   Platform Symphony MapReduce is the industry’s fastest job deployment architecture, with the ability to run MapReduce jobs on a distributed file system in sub-millisecond latency and data throughput of over 7,300 tasks per second. Back to top

5. Why is supporting multiple applications on a single distributed file system so important?

   Platform Symphony MapReduce supports up to 300 separate applications for MapReduce workloads running on a same distributed file system cluster.  Applications can run simultaneously while the Platform Symphony MapReduce distributed runtime engine manages the separate map and reduce tasks for each application on the cluster. This allows users tremendous flexibility in creating application dependencies, and it dramatically increases resource utilization while still maintaining a single management interface. In addition, it allows customers to leverage existing resources and maximize their IT infrastructure. Back to top

6. What are the limits of Platform Symphony MapReduce?

   Platform Symphony MapReduce offers 3 times scalability over its open source alternatives:

   • Up to 5,000 nodes and 40,000 cores per distributed file system
   • 40,000 concurrent tasks
   • 1,000,000 total tasks in a single job
   • 1,000 concurrent jobs with 300 MapReduce (Job Tracker) applications

   Additionally, Platform Symphony MapReduce is architected for virtually an unlimited number of priority levels for job scheduling, as opposed to Hadoop which can only support 5 levels. Back to top

7. If a customer does not use Platform Symphony MapReduce, what do they lose?

   Without Platform Symphony MapReduce, customers will pay more for their solution.  The added cost is reflected in both the need for more distributed file system servers, as well as additional overhead in the following ways:

   • Wasted, poorly utilized resources
   • Single failure points and business disruptions
   • Manual job restart and troubleshooting
   • Long cycle for job re-start
   • Lack of options – single application and file system lock-in
   • Poor performance, longer time to results
   • No sophisticated SLA capabilities, can’t meet business requirements from multiple lines of business
   Back to top

8. Is Platform Symphony MapReduce compatible with open source Hadoop stack?

   Yes, Platform Symphony MapReduce is compatible with open source Hadoop stack.  Platform Symphony MapReduce fully supports HDFS – Hadoop Distributed File System in the open source stack, as well open source APIs.  For customers wanting to use HDFS with Platform Symphony MapReduce, Platform Computing offers a support package for HDFS with Platform Symphony MapReduce. Back to top

9. How does Platform Symphony MapReduce compare to open source-based solutions?

   The current open source runtime engine lack enterprise-class capabilities required for production environments.  Because the core components of Platform Symphony MapReduce are based on Platform Computing’s nearly 20 years of workload management on distributed computing environments, Platform Symphony MapReduce immediately benefits from previous development.  Our technologies include powerful SLA-driven distributed workload engine, and resource orchestrator.  These components are both enterprise-class, proven technologies that are powering today’s leading companies with products such as Platform LSF and Platform Symphony. Built on the same core technologies,  Platform Symphony MapReduce is delivering  new capabilities to MapReduce applications. Back to top