Having fun with Google Cloud + RasPi

Having fun with Google Cloud + RasPi

91aeb42c5d9548918d1459f64240e503?s=128

Kazunori Sato

January 30, 2016
Tweet

Transcript

  1. 2.

    +Kazunori Sato @kazunori_279 Kaz Sato Staff Developer Advocate, Tech Lead

    for Data & Analytics Cloud Platform, Google Inc.
  2. 4.
  3. 6.

    Jupiter network 40 G ports 10 G x 100 K

    = 1 Pbps total CLOS topology Software Defined Network
  4. 7.

    Borg No VMs, pure containers Manages 10K machines / Cell

    DC-scale proactive job sched (CPU, mem, disk IO, TCP ports) Paxos-based metadata store
  5. 10.
  6. 11.

    Google BigQuery Demo: RegEx + GROUP BY on 10 B

    rows response read RegEx 10B ~10sec 372GB
  7. 13.

    BigQuery Analytic Service in the Cloud BigQuery Analyze Export Import

    How to use BigQuery? Google Analytics ETL tools Connectors Google Cloud BI tools and Visualization Google Cloud Spreadsheets, R, Hadoop
  8. 15.

    Blazingly Fast Capable of scanning 100B rows in ~20 sec

    Low Cost Storage: $0.020 per GB per month Queries: $5 per TB Fully Managed Use thousands of servers with zero-ops SQL Simple and Intuitive SQL with JS UDF Benefits of BigQuery
  9. 17.
  10. 18.
  11. 19.

    • Google Now • Google Photos • Gmail • YouTube

    • and more Google Brain: The Brain of Google services
  12. 20.
  13. 21.
  14. 24.

    24 Types of Detection • Label • Landmark • Logo

    • Face • Text • Safe search
  15. 25.

    25 Types of Detection Face Detection ◦ Find multiple faces

    ◦ Location of eyes, nose, mouth ◦ Detect emotions: joy, anger, surprise, sorrow Entity Detection ◦ Find common objects and landmarks, and their location in the image ◦ Detect explicit content
  16. 26.

    Making a request { "requests":[ { "image": { "content": "base64ImageString"

    }, "features": [ { "type": "LABEL_DETECTION", "maxResults": 10 }, { "type": "FACE_DETECTION", "maxResults": 10 }, // More feature detection types... ] } ] }
  17. 28.

    A new system for distributed, parallel machine learning: • Based

    on general-purpose dataflow graphs • Targeting heterogeneous devices ◦ single PC with CPU ◦ single PC with GPU(s) ◦ mobile device ◦ clusters of 100s or 1000s of CPUs and GPUs What is TensorFlow?
  18. 29.
  19. 30.

    Yet another dataflow systemwith tensors MatMul Add Relu biases weights

    examples labels Xent Edges are N-dimensional arrays: Tensors
  20. 33.