MityLytics for IoT Performance and Scalability Testing

/, Cassandra, Hadoop, IoT, Kafka, Spark, Storm/MityLytics for IoT Performance and Scalability Testing

MityLytics for IoT Performance and Scalability Testing

In speaking with folks running IoT stacks, which typically span several streaming and NoSQL technologies such as Kafka, Spark, Storm and Cassandra, it became apparent that they often struggle with understanding how an existing setup scales and performs with varying workloads.

So to address this, we enhanced our performance testing module to now offer IoT sensor emulation capabilities that allow you to test and understand end-user experience and platform behavior. The application profiles and characterizations that MityLytics derives are used as the basis for continuous performance optimizations/recommendations.

 

Figure 1 – Reference IoT Stack output to MityLytics

The MityLytics IoT test suite allows for end-to-end integration and performance testing of all components in the data pipeline such as Kafka, Storm/Spark, Hadoop and Cassandra shown in the IoT data pipeline figure above and understand infrastructure behavior.

To do so, MityLytics generates test loads with user defined characteristics such as:

  1. Message sizes
  2. Message volume
  3. Message rate

Figure 2 – Sample MityLytics Test Input

Messages can be generated using different stochastic models, which now makes it possible to:

  1. Profile and characterize data pipeline operations and performance to provide proactive remediation and ultimately automated performance management
  2. Identify hot spots in the data pipeline
  3. Identify weaknesses in the data pipeline with different rates of ingestion and processing
  4. Identify capacity constraints
  5. Report end-to-end message latencies
  6. Get system throughput

All the results are shown in easy-to-understand dashboards and tables.

Figure 3 – Sample MityLytics Output

Please contact us to learn more or to try this out for free.

By | 2017-10-26T12:37:23+00:00 October 23rd, 2017|Categories: Big Data Performance, Cassandra, Hadoop, IoT, Kafka, Spark, Storm|0 Comments

Leave A Comment