SMACK Stack - Data done Right #bbuzz
summary
SMACK combines Spark, Mesos, Akka, Cassandra and Kafka into pipelines that aim for second‑level reaction on large context rather than microsecond trading; Mesos auto‑schedules resources, Kafka provides append‑only buffering and replay days of data, Akka handles async decision trees (claims up to 50 million reqs per machine per second), and Cassandra targets ~10,000 reqs/machine/sec with append‑only denormalized writes. Used in automotive sensor evaluation and realtime advertising; Kafka replay rescues bad releases and Mesos scales consumers during backpressure.
AI-written summary · about
chapters (29)
- 0:00Intro
- 0:40SMACK Stack
- 1:27Spark
- 2:24SQL
- 5:20Acha
- 5:55Big Performance
- 6:48Cassandra
- 8:50Kafka
- 10:15Hadoop
- 11:13MapReduce
- 12:20High Frequency Processing
- 13:30The Sweet Spot
- 14:23The Juicy Parts
- 16:30Micro Batching
- 17:45RDDS
- 19:51Flexibility
- 22:28Reliability
- 23:20Reactive streams
- 24:45Back pressure
- 25:06Data flow
- 27:52Kafka is shit
- 28:24Cloud bare metal
- 29:49Using Kafka
- 31:40Stream Processing
- 32:14Buffering
- 33:21Reactive Streaming
- 35:38Scaling
- 37:01Scala
- 38:21Questions
related talks
28mKeynote Panel Discussion: SMACK Today and Tomorrow
19mKeynote Panel Discussion: SMACK in the Enterprise
59mStreaming Big Data w/Spark, Spark Streaming, Kafka, Cassandra & Akka
30mArchitecting Applications With Multiple Open-Source Big Data Technologies
39mThe Sparkling Solution that Scales to IoT
42mBuilding Robust Streaming Data Pipelines with Apache Spark
36mReal Time Big Data Analytics with Kafka, Storm, HBase #bbuzz
42mStreaming Data Pipelines on Apache Mesos: Lessons Learned