Distributed caching has reduced latency and now supports real-time stream processing
What is our primary use case?
My main use case for Hazelcast Platform is to build distributed cache systems across different services that we have. The biggest use is the caching. Additionally, from time to time, we use this for streaming and stream processing.
We use Hazelcast Platform to store user session states or session data, making it accessible across multiple application servers, allowing applications to scale horizontally. This is mostly clustering. We also use it for near-cache scenarios where we have to store frequently used data in the client application to reduce network latency.
What is most valuable?
I think the top core features of Hazelcast Platform for us are in-memory data storage, particularly the speed, because it can store data off-heap to eliminate long garbage collection pauses. The high-performance stream processing is also a significant feature. The processing engine runs directly where the data partitions live, eliminating network hops, which is very useful for us.
For our JVM-based services and applications, the in-memory data storage and stream processing features of Hazelcast Platform have made a difference for our team because we periodically used to pause the service to clean up deleted data, which would take seconds and crash the real-time processing and application. Hazelcast Platform bypassed this and stored the data directly in the server's off-heap native memory. Without this storage, we saw three-second garbage collection pauses, but now it is under two milliseconds. This is definitely an improvement. Another aspect is tiered storage; during peak seasons such as Black Friday, our systems have a lot of inventory updates. Hazelcast Platform retrieves it from the SSD without taking up memory space, which is extremely useful. In terms of stream processing, we used to have a server with the log, and another server was pulling the log from the network to analyze. Hazelcast Platform does all the analytics work inside the first server where the data is sitting, which eliminates all network overhead and enables real-time performance.
There have definitely been a lot more latency reductions and better SLA performance since using Hazelcast Platform, resulting in faster time to market overall for new features and capabilities because the architecture has become simpler. Developers can now focus on business logic rather than writing complex integration code due to this simplicity, and the time to market for delivery has increased.
Regarding improved SLAs or faster time to market with Hazelcast Platform, we had services that were handling about 25 million daily transactions. After implementing Hazelcast Platform, we were able to meet those SLA targets more consistently. It took time to migrate to Hazelcast Platform, but overall the SLAs were met and proved to be better than usual.
What needs improvement?
I think there are areas where Hazelcast Platform can improve, such as simplifying the cluster topology and sizing rules because they are still somewhat complex for someone new to Hazelcast. Understanding how the cluster topology forms and sizing rules work, such as partition balancing and traffic routing, should be much simpler. If one node has less RAM or a slightly slower CPU, it creates a cluster-wide performance bottleneck, which is critical, especially with transactional systems. Even though Hazelcast Platform has proven to be better, there can still be bottlenecks if the cluster topology and data partitioning are not easily understood. I also think that object handling and streamlined serialization should be prioritized; using standard Java serialization can be extremely slow. Providing native ultra-fast binary serialization out of the box, without requiring developers to write custom adapters, would be a significant improvement. It would be great if there were ready-to-use adapters for streamlined serialization and object handling.
For how long have I used the solution?
I have been using Hazelcast Platform for the past three years.
What do I think about the stability of the solution?
Hazelcast Platform is mostly stable.
What do I think about the scalability of the solution?
The scalability of Hazelcast Platform is decent.
How are customer service and support?
The customer support for Hazelcast Platform is good, with a lot of quality support available.
Which solution did I use previously and why did I switch?
We used traditional databases before switching to Hazelcast Platform. We encountered many problems with them, and Oracle Coherence was a key solution we used that had a high licensing cost and complex legacy management overhead. It is drastically easier to switch to Hazelcast Platform for deployment, especially on the Kubernetes side with its cloud-native architecture. The performance requirements are better compared to Oracle Coherence. We also had some Redis-based systems, but because of the multi-threaded nature of Hazelcast Platform, we switched some of those systems from Redis to Hazelcast Platform.
Which other solutions did I evaluate?
Before choosing Hazelcast Platform, we evaluated Redis and Oracle Coherence as two other major options where we already had some existing presence.
What other advice do I have?
My advice for others looking into using Hazelcast Platform is to study the architecture and data modeling thoroughly. Choosing the right topology for the application or platform is crucial when using Hazelcast Platform because it decouples your application lifecycle from Hazelcast Platform data cluster, allowing for updates to microservices without risking data loss. It is very important to consider the cluster sizes, serialization designs, and the system's availability and resiliency regarding metrics and eviction policies, such as time to live or max idle parameters when starting with Hazelcast Platform. I would rate this product an 8 out of 10.
Memory saver
What do you like best about the product?
I love the small memory of the library I embedded (since I have many issues with the memory of my pc).
It is also very fast and not complex at all and it saved my all my problems in the distributed systems for sure!
What do you dislike about the product?
Nothing to share about this...I would say as a con that it takes a bit more time in order ro coordinate yourself where and what is located
What problems is the product solving and how is that benefiting you?
Solved my problems with my relational databases where I was struggling with the whole backend team in managing correctly our db
Hazelcast is the one of the best tools for every day`s life
What do you like best about the product?
Hazelcast first of all is a really great distributed data grid. You can use distributed collections out of the box like Map, Set, Queue for microservices replicas to share data or for caching purposes. Also, it is possible to take distributed locks to prevent doing the same work simultaneously. The Streaming framework Hazelcast Jet has many needful features and is very fast. Can be run separately from services or as a library. Supports automatic rebalancing and replication. Multiple network options to make a cluster such as multicasting or kubernetes support.
What do you dislike about the product?
It`s a little bit tricky to set up cluster in Kubernetes but it`s a matter of reading the documentation carefully. You should write serializers for objects yourself but it`s your payment for the size of the memory footprint of your cache. Poor search options in in-memory data structures.
What problems is the product solving and how is that benefiting you?
Distributed cache, distributed locking for microservices architecture. Stream processing of the big data.
Recommendations to others considering the product:
Great tool for distributed caches and for the stream processing
Clustered in memory cache for your highly scalable application
What do you like best about the product?
We enjoyed using hazel cast cluster for one of the highly scalable SaaS applications serving 2 million transactions every minute
Hazel cast was very stable and we loved it
What do you dislike about the product?
We used Hazel cast back in 2013-2015 and those days I felt there were multiple opportunities to improve further. Some of those improvements which we had suggested didn't get implemented even after a a couple of years.
What problems is the product solving and how is that benefiting you?
We used it for reducing DB hits for our highly scalable SaaS application
Recommendations to others considering the product:
It used to be one of the well-known java based alternative for Memcached.
Give it a try:)
Simple and Powerful Stream Processing Engine
What do you like best about the product?
Performance and how it is easy to start implementation. In-memory processing is great adventage, because of high availability.
What do you dislike about the product?
Documentation is not as good as for Hazelcast IMDG and sometimes I do not know where I can find some information I am looking for.
What problems is the product solving and how is that benefiting you?
We needed Stream processing platform which allow to connect to Kafka and generate some output based on Python scripts provided by business analitycs. Jet allowed us to get messages from Kafka, translate them, enrich with data from software based on another Hazelcast IMDG, execute Python scripts and generate expected output.
Efficient in-memory database, best when used next to a relation database
What do you like best about the product?
I like Hazelcast being configurable to auto back-up features of it. When one of your hazelcast instance is down, remaining instances in the cluster recovers those data from backup and continue working where it left.
What do you dislike about the product?
I can't think of anything I don't like about Hazelcast IMDG, maybe the pricing would be the worst side of it.
What problems is the product solving and how is that benefiting you?
Hazelcast is solving our caching problem and fastens the data access time. It also makes it possible to spend less time on managing instances of cache. The instances can discover each other and communicate as default.
Recommendations to others considering the product:
I used Hazelcast embedded to the services. However, stand-alone usage of Hazelcast might provide a better/more sustainable cluster.
And Excellent Grid-Cache
What do you like best about the product?
Performance, ease-of-use, and JVM friendliness
What do you dislike about the product?
Hazelcast implementation requires application changes. However, solutions like the Heimdall Proxy can manage Hazelcast clusters by providing the caching and invalidation logic for Hazelcast.
What problems is the product solving and how is that benefiting you?
Improved response times and database scale via query caching
Recommendations to others considering the product:
The Heimdall Database Proxy provides the caching and invalidation for Hazelcast. You can create a SQL cache subsystem in minutes without any code changes.
High performance cache
What do you like best about the product?
The ability to have a distributed cache with very high performance numbers
What do you dislike about the product?
Setup on a distributed system can somethings be tricky, but with tools like Kubernetes it's much more simpler than used to be.
What problems is the product solving and how is that benefiting you?
We needed to serve multiple customers. Before, the only way to scale was adding a bigger machine, now, we can just add a new node.
Hazlecast IMDG helped us to reduce service transaction response times by an order of magnitude.
What do you like best about the product?
Hazlecast is great because it's distributed data-structures are extensions of commonly used java interfaces. Due to this our team was able to quickly grasp the framework, and implement our solution.
What do you dislike about the product?
At the time I was using this (2017), there weren't much documentation explaining the internal behavior, so we had to dig in to the source code.
What problems is the product solving and how is that benefiting you?
Hazlecast IMDG helped us to reduce service transaction response times by an order of magnitude, by allowing us to process at the data nodes, in memory. This is a huge cut down in network and disk I/O, when comparing to traditional architectures, where all the data is loaded to the service nodes. It also helped us increasing the number of CPUs available for a single transaction, thanks to it's distributed executor service.
More details - https://medium.com/@tharanga.hewa/distributed-computing-for-not-so-big-data-a7a14600d4b8
Recommendations to others considering the product:
May be the stateful architectures that Hazelcast presents are not for every situation, as they are hard to manage (talking about DevOps). Use it when there is a dying need for performance.
Hazelcast review
What do you like best about the product?
We had multiple services running independently. It was helping us to make financial data instantly available to all our services.
What do you dislike about the product?
It's costly and you cannt add huge data in this as it stores data in in-memory.
What problems is the product solving and how is that benefiting you?
Saved multiple back and forth calls between services as if you add data in any service and it will be instantly available to other services.
