Ira Palti’s Post

The blog post from ZITADEL https://lnkd.in/e9i_6kgB "Our decision to switch was also motivated by the need to improve latency and streamline infrastructure efficiency. CockroachDB's global distribution model, while really advantageous for high availability, introduced higher latency." It reminds us of the constraints of physics, precisely the speed of light. In the realm of decision-making processes that evolve the state of the system, understanding how your database manages consistency is crucial. The principle is straightforward: increased node coordination elevates latency while boosting availability. Conversely, reduced coordination decreases availability but also reduces latency. From the perspective of Command-Side operations, achieving both high availability and low latency (AP) is an illusion. Reasons why most Edge Databases in the Command-Side have very little practical application in reality (that does not mean it may work for you, or you made it work). #cqrs #distributedcomputing #eventualconsistency #database #edgecomputing #eventsourcing #eventdriven #softwarearchitecture #softwareengineering #systemdesign

Redis has joined the bandwagon of projects that dumped open-source software Licenses to focus more on profit. Another freemium disguised as OSS. What people didn't noticed is that Hazelcast made a stinky move 2 weeks ago: They kept the Apache 2.0 License, but practically speaking the community edition will be left to dry. There are lots of things in there they will NOT be maintaining anymore. They made it clear they want to focus on Enterprise. Hazelcast avoided the headlines by NOT changing it's License, but trust me, this refocus they are doing is a huge deal. You might want to start exploring options such as Infinispan & Apache Ignite. Take a look at the Blogpost from Hazelcast. #redis #hazelcast #oss #license

ScyllaDB Summit offers attendees exclusive access to our leadership and tech teams. At our most recent conference, Dor Laor and Avi Kivity fielded user questions on the new tablets replication algorithm, async shard-per-core architecture, tapping object storage with S3, the CDC roadmap, and comparing database deployment options. Here is what they had to say > https://lnkd.in/eMXBSiVy #ScyllaDB #NoSQL #database #opensource #techtips

🔥𝐂𝐚𝐜𝐡𝐢𝐧𝐠 𝐒𝐞𝐫𝐢𝐞𝐬 [3] In the previous post, we discussed cache: https://lnkd.in/dwjhcewv https://lnkd.in/d3xbhDGT 𝐃𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐞𝐝 𝐂𝐚𝐜𝐡𝐢𝐧𝐠 refers to caching mechanisms where the cached data is stored in a distributed, centralized store, accessible by all instances of your application. This is different from In-Memory Caching, which stores the cache in the memory of individual application instances, leading to potential inconsistency and duplication of cached data in environments with multiple servers or containers. It  useful in: - cloud-based environments where applications are deployed across multiple nodes. - microservice-based architectures where each service can share cached data. - scenarios requiring cache persistence after application restarts. Distributed caching allows your application to scale across multiple servers without duplicating data, ensures all instances share the same consistent data, and can be configured for high availability, making it resilient to failures. Some caches, like Redis, can even save data to disk so it persists after restarts. Disadvantages: - Network Latency:  Involves small latency due to network access, whereas in-memory caches are faster since they are stored locally. - Complex Setup - Additional Infrastructure Cost - Concurrency Handling: Requires mechanisms to handle concurrency issues when multiple instances attempt to update the same cached data. 📄 https://lnkd.in/dWMDj78F #dotnet #aspnetcore #systemdesign

September started with new cool updates 🔓 VictoriaMetrics Open Source: 🕵️♀️ New security features: Add release docker images built from scratch. These can be more preferable for use in environments with higher security standards.  vmagent and Single-Node #VictoriaMetrics: - Support for data ingestion via DataDog lambda extension aka /api/beta/sketches endpoint. - Reading Amazon CloudWatch metrics in OpenTelemetry format from Amazon Data Firehose - Service discovery support for Vultr. - Improved and optimized resource consumption while using stream aggregation. - vmctl: Support client-side #TLS configuration for VictoriaMetrics destination specified via cmd-line flags that used in #InfluxDB, Remote Read protocol, #OpenTSDB, #Prometheus, Promscale migration modes and for VictoriaMetrics native protocol. - vmauth: Improved CPU usage when proxying data ingestion requests Updated all dashboards to support Grafana 10+ and added new dashboard for vmauth Get more information in our changelog ➡️ https://lnkd.in/enw2NCzh

Discover the top 10 In-Memory Computing platforms revolutionizing data processing in 2024! Curious about which platforms made the cut? We've compared industry leaders like GigaSpaces XAP Skyline, Hazelcast, Apache Ignite, and more. Learn how these powerful tools can supercharge your operations and drive innovation. Don't miss our expert analysis and comparison table! Read the full article: https://lnkd.in/dR-R7xGk  Massimo Brignoli #InMemoryComputing #BigData #DataProcessing #TechInnovation #PerformanceOptimization

Move over Redis? Microsoft releases Garnet. The performance difference is pretty significant over Redis (nearly 100x in some scenarios). Written in C# and can extend with dotnet data structures. Can be used with existing Redis clients (mostly) because it relies on Redis serialization protocol for compatibility.

I suspect the code below ran out connection capacity, etc. Is there any interface I can send batch Check it out: https://lnkd.in/dS7xdzKN Join the conversation! #googlecloudstorage

🎉 🎉 🎉 A new version of AlloyDB Omni is out with numerous new features including cross-data-center replication that is essential for disaster recovery: https://lnkd.in/gmhrmxYc #google #database #portable #data #replication

Microsoft recently launched Garnet, a protocol-compatible Redis alternative (and then some). Want to use it right now? Easy, since it's protocol compatible with Redis, just use its standard impl of IDistributedCache (Microsoft.Extensions.Caching.StackExchangeRedis) as a 2nd level with FusionCache and you're good to go. Like, immediately today 🥳 https://lnkd.in/dJV3gdT7