Streaming Caffeine

Streaming Caffeine

By Jove ZhongTechnology
Download on the App Store

All Streaming Caffeine episodes

  • Streaming Caffeine E6: Giannis(Ververica) and Jove(Timeplus): Apache Flink, Pimon, CDC and Current23

    In this episode, Jove chatted with Giannis, Streaming Data Architect at Ververcia, the original creator of Apache Flink. 


    Giannis has been in the space for a loooong time: Pulsar/StreamNative/Flink/Aiven/Ververica. We talked about upcoming #Current23 (both of us will be there in person), #Flink #FlinkCDC #Debezium #Paimon #Database #DataWarehouse and many many more.


    (sorry there are some typo in the caption, such as Debezium. There is no way to edit the text in #Riverside, which I already raised the feature request via support)

    56 min
  • Streaming Caffeine E5: Leonid(Feldera) and Jove(Timeplus)

    Leonid, cofounder at Feldera, happened to be in Vancouver, to join VLDB, Very Large Database conference, and they won best paper award.

    Jove and Leonid met in Nero, a waffle shop, and talked about streaming processor, cloud, onprem, customers and many other topics.
    00:00 meet in waffle shop
    01:00 best paper in VLDB, incremental computation
    06:00 incremental computing in Timeplus
    07:00 challenges in incremental computing
    08:24 what SQL Feldera can run
    09:55, is Feldera, a stream processor or database?
    13:00, streaming analytics vs. continuous analytics
    14:30 about Jove and Timeplus
    17:57 why not Flink
    22:27 OSS
    24:22 scale out
    25:30 failover
    26:40 schema drift
    27:40 Timeplus streaming capabilities, join, time travel, unbounded query, bi integration
    34:00 how Feldera handles adhoc query
    37:12 customers
    40:50 cloud and onprem
    43:22 source/sink connectors, benthos and Debezium
    49:10 streaming app server

    53 min
  • Streaming Caffeine E3: Yaroslav(Goldsky) and Jove(Timeplus)

    Welcome to the latest episode of Streaming Cafe with your host, Jove!


    In this exciting interview, Jove is joined by a special guest, Yarosalv, Principal Software Engineer at Goldsky, to delve into the realm of data streaming, engineering challenges, and real-world experiences.


    Timestamps:

    00:00:00 Welcome Yarosalv00:01:40 Yaroslav self-intro, Activision, Shopify, Goldsky

    00:04:00 Yaroslav talk on Current23 and other talks he recommends00:14:20 Jove’s talk on Current23

    00:16:00 Yaroslav blog00:18:40 Blog writing on Friday00:21:00 data streaming, streaming ETL

    00:29:00 streaming vs batch, OLAP, CBO, JOIN, new challenges

    00:36:00 data democratization, real-time insights for everyone, not just executives

    00:36:55 data app vs data product00:39:10 next-gen AI is coming a data problem, Tesla e2e FSD v12

    00:41:00 blockchain is a distributed app platform, late events in web3

    00:50:00 tech community in Vancouver00:56:40 VLDB, Feldera, Materialize

    01:00:17 data streaming in depth is not easy. How about streaming databases


    Yaroslav's talk on Current23 and other talks he recommends

    * Dynamic Change Data Capture with Flink CDC and Consistent HashingKeynote

    * Restate: Stream Processing, but for Microservices

    * Isolating Streaming Ingest and Queries Using RocksDB (Rockset)

    * Deeply Declarative Data Pipelines (LinkedIn, Ryanne Dolan)

    * Off-Label Data Mesh: A Prescription for Healthier Data (Confluent, Adam Bellemare)

    * Learnings of Running Kafka Tiered Storage at Scale (Uber)

    * Evolution of Streaming Pipeline at Lyft (Lyft)* Why Serverless Flink Matters - Blazing Fast Stream Processing Made Scalable (Confluent)

    * Query Your Streaming Data on Kafka using SQL: Why, How, and What (Timeplus)

    1 hr 19 min
  • Streaming Caffeine E2: Tun(Quix) and Jove(Timeplus)

    Second episode of the new community driven video podcast: streaming caffeine. Talking about data streaming, data engineering, startup and coffee

    The talk happened on Aug 24, online, with Tun (VP of Data/DevRel at Quix) and Jove(cofounder and Head of Product at Timeplus)
    0:00:00 Intro of Tun and Quix
    0:01:00 About coffee
    0:04:00 Fast/slow coffee
    0:05:40 Life as VP of Data
    0:08:10 How Tun got to VP of Data
    0:10:30 Surf, wave, storm, streaming
    0:15:00 Snowboarding
    0:17:00 data streaming in Europ vs US
    0:27:00 Current23, Tun&Jove's talks
    0:31:00 Music on Current23
    0:33:00 Python and SQL
    0:38:00 Custom code and security
    0:39:50 Priority: who need what, when
    0:40:40 Cloud, BYOC, localhost
    0:50:20 Key personas, ICP
    0:51:30 Data Eng, Platform Eng, Data Analysts, Data Scienists
    0:52:00 ELT, ETL, EtLT
    0:59:00 Try new things, don't reinvent wheels
    1:00:00 Bubble tea,milk tea
    1:02:10 Quix +Timeplus
    1:02:52 SQL UDF, remote UDF, local UDF
    1:09:10 Coffee to Code (C2C)
    1:10:20 Wrap up. Streaming is cool. Talk to us
    1:12:20 batch + streaming. Choose what best fits you

    1 hr 15 min
  • Streaming Caffeine E1: Zander(Bytewax) and Jove(Timeplus)

    Very first episode of the new community driven video podcast: streaming caffeine. Talking about data streaming, data engineering, startup and coffee.


    The first talk happened on Aug 14, in Vancouver, with Zander (cofounder and CEO at Bytewax) and Jove(cofounder and Head of Product at Timeplus)

    1 hr 25 min