Hear Timeplus Enterprise updates from our Co-Founder and user cases from our customer. See the full recording of our Timeplus Live monthly webinar.
I am thrilled to announce that Timeplus Enterprise v2.6 is now Generally Available! This is the first GA release in 2025. In this release, we introduce a revolutionary Hybrid Hash Table to significantly reduce memory consumption for high cardinality data streams for joins and aggregations (for example, from 180GB to 20GB). This also enables more concurrent Materialized Views running in the Timeplus with reasonable costs. Other key enhancements include enhanced operational visibility, cross-deployment integration, as well as performance improvements and an enhanced user experience.
Powered by our lightweight, powerful, and efficient single-binary platform, Timeplus Enterprise offers flexible deployment options. Whether fully managed in the public cloud, self-hosted in your data center, running on your laptop, or even on edge devices, our platform adapts to diverse user needs and unblocks massive use cases, from handling large-scale workloads to operating in resource-contained edge environments, capturing real-time insights where data resides. Most importantly, Timeplus Enterprise drives down operational efforts and costs down with simplicity.
Our software has been rigorously tested and proven in a variety of industries and mission-critical use cases, including cybersecurity, algorithm-based trading, and real-time streaming pipeline and analytics. As we shared in a recent webinar, Timeplus partnered with RocNet Supply to monitor network telemetry data with SQL-based processes and Grafana visualization.
"Using Timeplus and Grafana together has been awesome! Timeplus simplifies what was a delicate manual transform process into an automatic SQL-based process."
Jason Patterson | Director of Network Architecture, RocNet Supply
Today, we are excited to upgrade our offerings with the release of Timeplus Enterprise v2.6. This unified, enterprise-grade product is available through our fully managed cloud service or can be installed on various infrastructures, including bare metal, Docker, and Kubernetes.
Key Breakthroughs in Timeplus Enterprise v2.6
The v2.6 release of Timeplus Enterprise introduces several groundbreaking features:
Revolutionary hybrid hash table for join and aggregations
Enhanced operational visibility
Performance improvements
Improved data management
Enhanced user experience
Revolutionary Hybrid Hash Table
Imagine that you are running an analytics workload for a large e-commerce platform, such as Shopifly. There can be hundreds of millions of product SKUs that change slowly. On the other hand, every second there are many new events in the orders transaction table. For each new order, you need to look up the products dimension table to enrich the data. This sounds easy but is a hard problem to solve: you don’t want to slow down the OTLP database, so the data should be replicated to an OLTP database or Kafka topics, but streaming JOINs with such high cardinality would easily lead to out-of-memory errors.
Another scenario is blockchain analysis. One of Timeplus’ customers offers a platform for end users to index and analyze blockchain data. There are 2.6 billion unique transactions, 1.9 billion transfers and 2.1 billion actions. When the data is revisioned during blockchain reorganization or any update to a single transaction, the platform needs to show the revised aggregation results within one or two seconds, instead of reprocessing billions of data records in the chain.
In such scenarios, the default in-memory hash table for table JOIN can easily consume a lot of system memory. For example, one of our customers observed 180GB memory consumption with 10 materialized views in Timeplus to process such a massive scale of data. With the new Hybrid Hash Table in Timeplus Enterprise 2.6, which uses both the memory and the local disk to store the internal state as a hash table, only 20GB of memory is consumed, while still showing real-time streaming aggregation results every second.
SELECT account_id, sum(total) AS total
FROM transfers
GROUP BY account_id
EMIT PERIODIC 1s ON UPDATE SETTINGS default_hash_table=’hybrid’;Enhanced Operational Visibility
Gain complete transparency into your system's performance through comprehensive monitoring of materialized views and streams. Track state changes, errors, and throughput metrics via system.stream_state_log and system.stream_metric_log.
Those metrics are available as Prometheus endpoints, and you can also query them using SQL.
For example, the following SQL will show the read/write throughput for each data stream in Timeplus for the past 5 minutes.
SELECT
name,
avg((read_bytes / elapsed_ms) * 1000) AS read_bps,
avg((read_rows / elapsed_ms) * 1000) AS read_eps,
avg((written_bytes / elapsed_ms) * 1000) AS written_bps,
avg((written_rows / elapsed_ms) * 1000) AS written_eps
FROM table(system.stream_metric_log)
WHERE elapsed_ms > 0 and _tp_time > now()-5m
GROUP BY name;Performance Improvements
In this release, we pushed our unparalleled performance to a new level. One of our customers needs to write billions of rows to ClickHouse from Timeplus. This is done via a materialized view with a ClickHouse external table as the target, for example:
--DDL in ClickHouse
CREATE TABLE default.product
(
`i` Int32,
`j` String,
`_tp_time` DateTime64(3)
)
ENGINE = MergeTree
ORDER BY i
SETTINGS index_granularity = 8192;
--DDL in Timeplus
CREATE EXTERNAL TABLE ch_product_ext
SETTINGS type = 'clickhouse', address = 'clickhouse:9000', table = 'product';
CREATE MATERIALIZED VIEW mv INTO ch_product_ext AS
SELECT i, j, _tp_time FROM rand_data;With Timeplus Enterprise v2.5, the write throughput is approximately 3 million rows per second. This could be measured by running the following SQL in ClickHouse:
clickhouse :) select count(1)/DateDiff(s,min(_tp_time),max(_tp_time)) from product;
SELECT count(1) / dateDiff('second', min(_tp_time), max(_tp_time))
FROM product
┌─divide(count(1), dateDiff('second', min(_tp_time), max(_tp_time)))─┐
1. │ 2971131.1861743913 │ -- 2.97 million
└────────────────────────────────────────────────────────────────────┘
1 row in set. Elapsed: 3.268 sec. Processed 3.78 billion rows, 30.26 GB (1.16 billion rows/s., 9.26 GB/s.)As you can see, there are a total of 3.78 billion rows inserted into ClickHouse via Timeplus. By comparing the minimal and maximal event timestamp, we can figure out the write throughput is approximately 3 million per seconds.
In Timeplus Enterprise 2.6, we enhanced the connection pooling and other mechanisms. The same SQL pipeline can achieve 3.6 million per second throughput. We can further tune the insert block size and batching strategies, and make it 2 times faster, i.e. 6 million rows per second.
CREATE MATERIALIZED VIEW new_mv INTO ch_products_ext AS
SELECT i, j, _tp_time FROM rand_data
SETTINGS max_insert_block_size=130000, max_insert_block_bytes=102400, insert_block_timeout_ms = 1;
Improved Data Management
In this release, you can add new columns to an existing stream using:
ALTER STREAM stream_name ADD COLUMN column_name data_typeYou can also truncate the data in a stream without recreating it:
TRUNCATE STREAM stream_nameTimeplus Enterprise 2.6 also supports multiple database namespaces. You can organize your streams, materialized views, and other objects in different databases:
CREATE DATABASE IF NOT EXISTS my_database;
USE my_database;
CREATE STREAM my_stream (raw string);
INSERT INTO my_stream(raw) VALUES ('hello');
SELECT * FROM table(my_stream);
--if you are using other databases
SELECT * FROM table(my_database.my_stream);Last but not least, you can create a materialized view and use Timeplus External Stream as the target. This allows you to continuously write data to a remote Timeplus Enterprise or Timeplus Proton after stream processing.
UI Improvements
We start seeing some of our customers create thousands of materialized views or ClickHouse External Tables. In this version, we added a shortcut to quickly access any objects in Timeplus. It’s Command + K on macOS and Control+K on PC.
We have also taken feedback from our community and redesigned the SQL Console. The updated layout allows you to write longer SQL statements and quickly access to the stream schemas, common functions, query history, or bookmarks.
We have also added new UI wizards for Coinbase and Apache Pulsar integration.
The demo data generator is now powered by the native Timeplus Random Stream. You can customize the SQL to adjust the events-per-second or customize the data generation logic.
Enhancement for Grafana Plugin for Timeplus
Grafana has long been one of the most popular tools for real-time monitoring and data visualization, helping organizations track metrics and analyze trends through a single, user-friendly interface.
Today, along with the release of Timeplus Enterprise v2.6, we’re thrilled to announce the release of Timeplus Grafana Plugin v2.1, with the following enhancements:
Show the SQL syntax error and recommendations for next step in the Grafana panel
Updated the batching strategies to improve performance of rendering large amounts of results via streaming queries.
Support query variables, annotations and alerts.
Jason Patterson, Director of Network Architecture at RocNet Supply, demonstrates how he uses the latest Grafana plugin to apply network streaming telemetry. Watch the webinar recording to learn more.
Learn More
To explore the capabilities of Timeplus Enterprise v2.6, please check out https://docs.timeplus.com/enterprise-v2.6 for installation guide as well as change logs.
Thank you for being a part of our journey. We look forward to your continued support and collaboration as we move towards a future powered by real-time streaming analytics.
Ready to try Timeplus Enterprise? Try for free for 30-days.
Join our Timeplus Community! Connect with other users or get support in our Slack community.