Google Cloud Dataflow Reviews

Google Cloud TPU Reviews

Google Cloud Dataflow mentions (14)

• How do you implement CDC in your organization

  Imo if you are using the cloud and not doing anything particularly fancy the native tooling is good enough. For AWS that is DMS (for RDBMS) and Kinesis/Lamba (for streams). Google has Data Fusion and Dataflow . Azure hasData Factory if you are unfortunate enough to have to use SQL Server or Azure. Imo the vendored tools and open source tools are more useful when you need to ingest data from SaaS platforms, and... Source: over 1 year ago

• Here’s a playlist of 7 hours of music I use to focus when I’m coding/developing. Post yours as well if you also have one!

  This sub is for Apache Beam and Google Cloud Dataflow as the sidebar suggests. Source: over 1 year ago

• How are view/listen counts rolled up on something like Spotify/YouTube?

  I am pretty sure they are using pub/sub with probably a Dataflow pipeline to process all that data. Source: over 1 year ago

• Best way to export several GCP datasets to AWS?

  You can run a Dataflow job that copies the data directly from BQ into S3, though you'll have to run a job per table. This can be somewhat expensive to do. Source: over 1 year ago

• Why we don’t use Spark

  It was clear we needed something that was built specifically for our big-data SaaS requirements. Dataflow was our first idea, as the service is fully managed, highly scalable, fairly reliable and has a unified model for streaming & batch workloads. Sadly, the cost of this service was quite large. Secondly, at that moment in time, the service only accepted Java implementations, of which we had little knowledge... - Source: dev.to / about 2 years ago

Google Cloud TPU mentions (5)

• Pathways Language Model (Palm): 540B Parameters for Breakthrough Perf

  According to https://cloud.google.com/tpu, each individual TPUv3 has 420 Teraflops, and TPUv4 is supposed to double that performance, so if that guess is correct, it should take a few seconds to do inference. Quite impressive really. - Source: Hacker News / about 2 years ago

• The AI Research SuperCluster

  You can also rent a cloud TPU-v4 pod (https://cloud.google.com/tpu) which 4096 TPUv-4 chips with fast interconnect, amounting to around 1.1 exaflops of compute. It won't be cheap though (excess of 20M$/year I believe). - Source: Hacker News / over 2 years ago

• Stadia's future includes running the backend of other streaming platforms, job listing reveals

  Actually, that's done with TPUs which are more efficient: https://cloud.google.com/tpu. Source: almost 3 years ago

• Nvidia CEO: Ethereum Is Going To Be Quite Valuable, Transactions Will Still Be A Lot Faster

  TPU training uses Google silicon and is thus a true deep learning alternative to Nvidia. Source: almost 3 years ago

• Server Question

  The server choice really depends on how much CPU and RAM the requests take, how many users will be hitting the server, etc. You can start with a $5/month Digital Ocean server (or AWS or Google) and see if that works for you. Or you can outsource the server administration to Amazon or Google if you don't want to deal with it or need specialized tpu hardware. Source: about 3 years ago