Scaling to Millions of Devices and Billions of Events

467 views
338 views

Published on

Join us to learn how Salesforce Platform R&D is exploring new ways to scale data processing and aggregation from millions of devices. Learn about new frameworks, the challenges with large number of data sources and volumes of information, and making it all work. This session covers internal projects under incubation.

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
467
On SlideShare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
9
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Scaling to Millions of Devices and Billions of Events

  1. 1. Scaling to Millions of Devices and Billions of Events Oleg Gusak, Salesforce.com, LMTS, Performance Engineering
  2. 2. Safe harbor Safe harbor statement under the Private Securities Litigation Reform Act of 1995: This presentation may contain forward-looking statements that involve risks, uncertainties, and assumptions. If any such uncertainties materialize or if any of the assumptions proves incorrect, the results of salesforce.com, inc. could differ materially from the results expressed or implied by the forward-looking statements we make. All statements other than statements of historical fact could be deemed forward-looking, including any projections of product or service availability, subscriber growth, earnings, revenues, or other financial items and any statements regarding strategies or plans of management for future operations, statements of belief, any statements concerning new, planned, or upgraded services or technology developments and customer contracts or use of our services. The risks and uncertainties referred to above include – but are not limited to – risks associated with developing and delivering new functionality for our service, new products and services, our new business model, our past operating losses, possible fluctuations in our operating results and rate of growth, interruptions or delays in our Web hosting, breach of our security measures, the outcome of any litigation, risks associated with completed and any possible mergers and acquisitions, the immature market in which we operate, our relatively limited operating history, our ability to expand, retain, and motivate our employees and manage our growth, new releases of our service and successful customer deployment, our limited history reselling non-salesforce.com products, and utilization and selling to larger enterprise customers. Further information on potential factors that could affect the financial results of salesforce.com, inc. is included in our annual report on Form 10-K for the most recent fiscal year and in our quarterly report on Form 10-Q for the most recent fiscal quarter. These documents and others containing important disclosures are available on the SEC Filings section of the Investor Information section of our Web site. Any unreleased services or features referenced in this or other presentations, press releases or public statements are not currently available and may not be delivered on time or at all. Customers who purchase our services should make the purchase decisions based upon features that are currently available. Salesforce.com, inc. assumes no obligation and does not intend to update these forward-looking statements.
  3. 3. Oleg Gusak Lead Member of Technical Staff Performance Engineering
  4. 4. Outline ▪ Introduction: Internet of devices ▪ Architecture of the devices framework ▪ Challenges for Performance Testing ▪ Architecture of the Test System ▪ Results and Q&A
  5. 5. Internet of Devices
  6. 6. Internet of Devices Billions devices produce massive amount of data
  7. 7. Internet of Devices Not every message carries important information
  8. 8. Internet of Devices Transform data to actionable information
  9. 9. Devices Data Your org in Salesforce.com API Actionable information
  10. 10. Architecture
  11. 11. Challenges of Performance Testing • 1 Million devices • 1 Million events per second • Ad-hoc setup
  12. 12. Key components of the solution • Automate performance test end to end
  13. 13. Key components of the solution • Reuse orchestration developed for internal performance testing
  14. 14. Key components of the solution • Non-blocking IO http client
  15. 15. Key components of the solution • Amazon elastic cloud and its API
  16. 16. Key components of the solution • Sumo Logic for log analysis
  17. 17. Rest API Storm Kafka Storm Load VM Kafka Rest API Architecture of the test system Load VM Results processor Load VM REST API of cluster manager ELB Controller Sumo Logic
  18. 18. Architecture of the test system • Adopted internally developed automation tool Suzuki for testing in EC2
  19. 19. Suzuki - load test orchestrator Load Collect results Load generator Log manager Parts Setup Stages
  20. 20. Architecture of the test system • Common image for load VMs tuned to support large number of concurrent connections
  21. 21. Architecture of the test system • Load VMs are managed via Amazon API
  22. 22. Architecture of the test system • Metadata of the target system under test is retrieved via REST API of the cluster manager
  23. 23. Workflow of a performance test Provision load instances Load VM Load VM Controller Load VM Amazon API
  24. 24. Workflow of a performance test Configure load instances Load VM Load VM Controller Load VM Amazon API REST API
  25. 25. Workflow of a performance test Rest API Storm Storm Load VM Kafka Load VM Kafka Controller ELB Load VM Rest API Generate load
  26. 26. Workflow of a performance test Rest API Storm Storm Load VM Kafka Controller ELB Load VM Kafka Rest API Collect logs and process results Load VM Sumo Logic
  27. 27. Single REST API instance
  28. 28. 80,000 requests per second, 10 c1.xlarge load instances
  29. 29. 80,000 requests per second, 10 c1.xlarge load instances
  30. 30. Oleg Gusak LMTS, Performance Engineering

×