Tech Talks
The monthly tech talks is to bring the Phoenix community together to have technical discussions, share knowledge, and initiate further collaborations, innovations and improvements.
The topics can be any technical subject related to Phoenix, including its internals, interfaces, operational aspects, and use cases, and technologies that it leverages, and can possibly leverage or adapt.
When: First Thursday of each month at 9AM PST
Duration: 90 minutes (to allow the audience to participate and ask questions)
Meeting Link: To attend the meetings, please use the video conference link to be provided here
Suggest Topics: To suggest a topic to present in an upcoming meeting, please send an email to the user and dev list
Upcoming Tech Talks
| Title | Presenters | Abstract | When |
|---|---|---|---|
| TBD |
Previous Tech Talks
| Title | Presenters | Abstract | Resources | Recording | When |
|---|---|---|---|---|---|
| JSON Support for Apache Phoenix | Gokcen Iskender | JSON is a popular data format that is used for exchanging data in many applications. Many RESTful services return or accept data formatted as JSON. Also several databases support JSON as their primary data format. This talk proposes changes which will introduce JSON support in Phoenix to better support our customers. | Slides | Recording | April 07, 2022 |
| Online Data Format Change in Phoenix | Gokcen Iskender | Phoenix lets the users to make certain changes to the table/index schemas via ALTER TABLE and ALTER INDEX. Some schema changes such as changing the row key (primary keys), the type of a column, the table storage format, the column encoding, etc. requires table/index rows to be re-written. This talk introduces Online Data Format change feature which will let Phoenix users to make those changes with no service interruption. The focus will be on changing column encoding and table storage format. | Slides | Recording | Oct 07, 2021 |
| Phoenix High Availability | Mingliang Liu, Daniel Wong, and Abhishek Singh Chouhan | This talk introduces Phoenix High Availability (HA) feature which allows Phoenix users to interact with multiple Phoenix/HBase clusters in order to achieve additional availability compared to a single cluster. In particular we target the common deployment configuration of two HBase clusters with master/master asynchronous replication enabled between the queried tables, but with consideration to future extensions in use cases, replication, and number of clusters. | Slides | Recording | July 01, 2021 |
| Atomic Upserts in Phoenix | Tanuj Khurana | Atomic upserts are supported for tables without global indexes. The current design of atomic upserts cannot be expanded to support tables with global indexes since doing index table updates results in cluster wide deadlocks. This talk covers how atomic upserts can be used, details around how it is implemented today, its limitations and how we can leverage the new global secondary indexing design to support atomic upserts on tables with global indexes. | Slides | Recording | May 06, 2021 |
| Phoenix Query Server, PhoenixDB, and Hue | Josh Elser, Istvan Toth, and Romain Rigaux | PhoenixDB is a new project which is a Python library for interacting with Phoenix with Phoenix Query Server (PQS). Hue is an open source SQL assistant for databases and data warehouses. In this talk, we give an overview of PQS, and recently completed and planned improvements for it, and introduce the PhoenixDB project. Then, we present Hue, and touch on the integration of PhoenixDB into Hue. Finally, we present a live demo of using PhoenixDB within Hue. | PQS and PhoenixDB Slides - Hue Slides | Recording | April 01, 2021 |
| Strongly Consistent Global Secondary Indexes | Kadir Ozdemir | Global secondary indexing, which enables efficient queries on non-primary key fields, is central to many use cases. Phoenix implements global secondary indexing by pairing a data table (the table that holds the entire user data) with physically separate (index) tables that are indexed on a different set of columns. Some use cases demand strong consistency between the data table and its indexes, meaning that regardless of whether a given query is served from a data table or an index table, the same result is returned. This talk presents the secondary indexing design to meet this strong consistency requirement. | Slides | Recording | March 04, 2021 |