Aerospike Standup Volume 49

Welcome to the Aerospike Standup. Here are some highlights:

We are restarting developer meetups! The first one is scheduled on Tuesday, July 19th in our Mountain View office. Join us if you can and learn from experts at PayPal about robust and scalable monitoring of over 2000 Aerospike nodes as well as the high level architecture of their AI/ML-based real-time fraud detection and prevention system.

We released Aerospike Monitoring version 2.0.0 adding support for many new Aerospike 6.0 metrics in the Grafana dashboards. (Note that the 2.0.0 Grafana dashboards are not backwards compatible with Aerospike servers older than 6.0.0.0.) We also released Aerospike Prometheus exporter version 1.6.0 to support metrics added in Aerospike server 6.0.

The blog post Are Graph Databases Finally Ready for Prime Time? announces the Firefly Alpha Program, the code name for Aerospike’s graph database offering. Firefly combines Aerospike Database with Apache TinkerPop™. If you would like to sign up, please reach out at firefly-alpha@aerospike.com.

The article Data Modeling for Speed At Scale (Part 2) continues where last month's post left off, and describes how applications can optimize speed and scale by leveraging Collection Data Types (CDTs).

See below for details on the blog posts, releases, questions from the community, and events collected from this past month.

Upcoming Events

• Aerospike Scale Warriors Meetup - Featuring PayPal.

  Register now!

New Blog Posts

• Data Modeling for Speed At Scale (Part 2)

  by Neel Phadnis

• Are Graph Databases Finally Ready for Prime Time?

  by Ishaan Biswas

Releases

• Java Client

  6.0.1

• Go Client

  6.2.0, 6.1.0, 5.9.0

• Node.js Client

  5.0.2, 4.0.4

• C Client

  6.1.0

• Ruby Client

  2.21.1, 2.21.0

• Connect for Spark

  3.4.1, 3.3.1_spark3.1, 3.2.2

• Monitoring Stack

  2.0.0

• Prometheus Exporter

  1.6.0

• Tools

  7.0.5, 7.0.3

Questions From the Community

• Nodes in different clusters on the same subnet with mesh configuration form ONE cluster

• Max-write-cache in pmem (persistent memory)

• Aerospike connector for presto

• Proto-fd-max

• Difference between batch reads/writes and latency read/write ops_per_sec from a monitoring perspective

• Capacity for in memory storage

• Adding a new node does not decrease the system memory usage

• Presence in user-path does not suffice for udf registration

• Node.js client upgrade from 1.0 to 3.5

• Namespace configuration in Aerospike 5 - namespace vs dc contexts

• Why insert/put within loop insert only 1 row?

• ConnectionError: (-10L, 'Socket read error: 104, :3000, 35382', 'src/main/aerospike/as_socket.c', 248, False)

• Interaction of List operation (getByValue) and more complex data types

• C#, GetByValue is not working

• Deploy Aerospike with Ceph

• In which situation one record may have multi generations in storage

• Debian 11 package for Aerospike Server

• Migrate Data Model from Scylla to Aerospike

• Change/modify replication-factor on the fly? (AER-3242)

• Bulk/Batch Updates (AER-6499)

• Batch Remove (AER-6499)

• Batch Writes (AER-6499)

• Write-block-size configuration parameter exact meaning

• What is the lowest compatible Aerospike Enterprise Server version for Java client 5.1.11?

• Write-q stuck at max

• Hot key errors even with 'memory' namespace

• Can we change the time one node take to join cluster after restart?

• Atomic increment via operate command using java client (5.1.5.1) fails