![\"EloqCloud](\"https://www.eloqdata.com/assets/images/eloqcloudforeloqdoc-ce6e460607481d818820d3643a64e56e.jpg\")
After the successful launch of EloqCloud for EloqKV, today we are excited to announce EloqCloud for EloqDoc — our new MongoDB-compatible document database built for the cloud era.
\nEloqCloud for EloqDoc brings a breakthrough architecture that redefines the balance between cost, scalability, and performance for document workloads.
\nEloqCloud for EloqDoc introduces a production-grade free tier with dedicated resources:
\nYou can start building your application with a real, isolated and predictable document database — not shared compute or ephemeral sandboxes.
\nBelow is the performance comparison between MongoDB Atlas and EloqCloud for EloqDoc (Free Tier) across three typical workloads — Read Only, Read Write, and Write Only.
\nThe results show a dramatic difference: EloqCloud for EloqDoc consistently outperforms MongoDB Atlas by nearly an order of magnitude, delivering up to 10× higher throughput even on its free-tier environment.
\n![\"EloqCloud](\"https://www.eloqdata.com/assets/images/eloqdocmongofreetier-8b3d33ca67c743143cac44a2ac9f2802.jpg\")
Beyond the industry’s first production-grade free tier, EloqCloud for EloqDoc offers the most flexible and affordable paid plans in the market.
\nYou pay only for what you use — with no forced replication, no bundled resources.
| Resource Type | Unit Price | Notes |
|---|---|---|
| Storage | $0.10 / GB / month | Includes cross-AZ high availability |
| Compute | $0.14 / vCPU / hour | Independent compute billing |
You can scale compute and storage independently — keep a large dataset with a small compute footprint, or burst compute when you need it.
\nBelow is the price comparison between MongoDB Atlas and EloqCloud for EloqDoc under the same configurations.
\nWhile both platforms offer similar compute and memory setups, the difference in architecture leads to dramatically different pricing.
\nEloqCloud for EloqDoc leverages object storage as its primary data layer and local SSDs for caching, delivering significant cost savings compared to MongoDB Atlas, which depends on multiple replicas and EBS-based storage.
\n![\"EloqCloud](\"https://www.eloqdata.com/assets/images/mongoeloqdocprice-2ee4068a0225e27a9339550a236c419e.jpg\")
How can EloqCloud for EloqDoc deliver a production-grade free tier and significant cost savings compared to MongoDB Atlas? The answer lies in its Cost-Effective Architecture, a reimagined foundation purpose-built for the cloud.
\nMongoDB Atlas requires at least three replica set nodes to maintain availability — which triples compute and memory consumption. Even worse, it relies on EBS volumes as the default storage layer, which are significantly more expensive than cloud object storage. Since each EBS volume maintains three replicas per compute node, a standard three-node replica set ends up with nine total storage replicas, leading to excessive storage overhead and cost.
\nIn contrast, EloqCloud for EloqDoc achieves the same or better reliability with a storage-centric design that eliminates redundant replicas and maximizes efficiency:
\nThis architecture enables EloqCloud for EloqDoc to deliver lower TCO and superior performance for real-world production workloads.
\nEloqCloud for EloqDoc is now available for public access.
\nStart for Free and experience the next-generation document database —
\nMongoDB-compatible, object storage first, high performance and cost effective.
We’re excited to announce Public Access for the EloqCloud platform!\nThe first product EloqCloud for EloqKV — a key-value database with Redis-compatible API — officially joins the EloqCloud family of cloud-native SaaS databases.
\n![\"EloqCloud](\"https://www.eloqdata.com/assets/images/launcheloqcloud-753981a5032bd0627294dab5ca7cc92a.png\")
EloqCloud is the SaaS platform for EloqDB, designed to deliver an economical, scalable, and predictable cloud experience for developers — using the APIs they already love.
\nEloqCloud is a product matrix, supporting a variety of data models and compute APIs, including SQL, Redis, MongoDB, Vector, Graph, as well as an all-in-one solution, ConvergedDB, that support operation cross data models. These product features will be rolled out over time — and today, we're thrilled to introduce the first product EloqCloud for EloqKV — a key-value database with Redis-compatible API.
\nEloqCloud for EloqKV is a cloud-native transactional key-value database with full Redis API compatibility.
\nIt goes far beyond a traditional in-memory cache — offering true ACID compliance database, built for durability, availability, and performance. Writes are replicated before acknowledging to the client with cross-AZ persistence to withstand Availability Zone failures, while the performance matches existing in-memory cache solutions.
\nHere’s a closer look at its key capabilities:
\n1. Scale to Zero
\nEloqCloud for EloqKV supports scale to zero. When idle for a short period of time, it can automatically free up all compute resources — and save costs. No usage? No bill.
2. Dedicated Resources
\nChoose the SKU that fits your workload. Resources are dedicated, ensuring predictable performance without noisy neighbors.
3. Economic High Availability
\nWhy pay for 2x or 3x replicas just to achieve high availability?
\nWith EloqCloud for EloqKV, one compute replica is enough. Our Data Substrate architecture decouples compute, memory, log and storage, enabling:
Save big — without sacrificing reliability.
\n4. Hybrid Storage Architecture
\nSay goodbye to overpaying for cloud Elastic Block Storage (EBS) or risking instance local SSD data loss.
5. High Performance
\n6. Full ACID Transactions
\nUnlike typical Redis-like services, EloqCloud for EloqKV supports real ACID transactions, with familiar Begin / Commit / Rollback semantics — enabling SQL-style transactional patterns over Redis API.
7. Seamless Scalability
\nSelect the SKU that best matches your workload.\nAuto-scaling is on our roadmap to make dynamic scaling effortless.
Ready to experience the future of cloud databases?
\nApply For Access today — and receive an exclusive invitation to try EloqCloud for free.
![\"Building](\"https://www.eloqdata.com/assets/images/joinwaitlist-01717d235834590d0a4fe393ad2a927c.png\")
📺 Watch our quick demo video below to see EloqCloud in action!
\nWe'll be sharing detailed benchmark results and comparisons with other key-value stores in upcoming posts — stay tuned!
", "url": "https://www.eloqdata.com/news/2025/04/29/eloqcloud", "title": "EloqCloud for EloqKV: Public Access Now Open", "summary": "Announcing the first key-value product on the EloqCloud platform — a cloud-native, transactional database fully compatible with the Redis API.", "date_modified": "2025-04-29T00:00:00.000Z", "author": { "name": "EloqData" }, "tags": [ "Product" ] }, { "id": "https://www.eloqdata.com/news/2025/03/11/eloqsql-eloqdoc", "content_html": "We are thrilled to introduce EloqSQL and EloqDoc, two groundbreaking additions to our database portfolio.
\nEloqSQL is a MySQL-compatible, fully distributed relational database, while EloqDoc is a MongoDB-compatible, distributed JSON document database. Both are built on our modular Data Substrate technology, inheriting all the benefits of this cutting-edge architecture, including best-in-class performance in all scales and full distributed ACID transactional capabilities with selectable isolation levels to ensure data integrity and reliability.
\nEloqSQL is a MySQL-compatible distributed relation database that sets a new standard for high performance, scalability, compatibility, and resilience. Designed for modern applications, it supports multiple writers, secondary indexes, cross-AZ data replication, and hot-backup for non-stop operation. Engineered to handle the most demanding workloads with ease, EloqSQL transforms the way you manage and process data with the most widely accepted API in the industry.
\nEloqDoc is a MongoDB-compatible distributed JSON document database that can leverage tiered cloud storage for unparalleled flexibility and efficiency. With fast and dynamic capacity scaling, EloqDoc adapts seamlessly to changing demands, ensuring optimal performance at all times. It delivers significantly better performance than MongoDB in many scenarios while maintaining API compatibility and a painless migration path. EloqDoc’s unprecedented distributed transaction capability eliminates one of the major distinctions between traditional NoSQL and SQL databases, combining the best of NoSQL technology with the power of cloud storage for an exceptional data management experience.
\nBoth EloqSQL and EloqDoc are immediately available for preview and can be downloaded from our website. You can learn about how to set up a single node test server or a cluster following our documents.
\nTogether, EloqSQL and EloqDoc represent a significant leap forward in database technology, providing robust, scalable, and high-performance solutions for a wide range of applications. We invite you to explore these new tools and experience the future of data management with us. Stay tuned for more updates and in-depth insights into how these products can simplify your data infrastructure and reduce your database costs.
\nLooking ahead, we are excited about the continued expansion of the Eloq family of databases. The Data Substrate architecture gives us unparalleled flexibility and agility to rapidly develop high-quality, competitive products that meet the evolving data needs of modern AI-driven applications.
", "url": "https://www.eloqdata.com/news/2025/03/11/eloqsql-eloqdoc", "title": "Announcing Our Latest Innovations - EloqSQL and EloqDoc", "summary": "Announcing two new products - MySQL API compatible relational database EloqSQL and MongoDB compatible JSON document database EloqDoC.", "date_modified": "2025-03-11T00:00:00.000Z", "author": { "name": "EloqData" }, "tags": [ "Product" ] }, { "id": "https://www.eloqdata.com/news/2025/03/10/monster-scale-summit", "content_html": "Redis is widely used for caching and real-time data processing, but its scalability limitations create significant challenges for modern applications. At Monster Scale Summit 2025, we introduced EloqKV, a high-performance, distributed transactional database that overcomes Redis' scaling bottlenecks by leveraging ScyllaDB. This blog post explores how EloqKV enhances scalability, performance, and transaction capabilities while maintaining Redis API compatibility.
\nAt Monster Scale Summit 2025, Eloqdata is honored to be the featured speaker, presenting how EloqKV on ScyllaDB, our high-performance, distributed transactional database, tackles Redis scaling bottlenecks. Hosted by ScyllaDB, a distributed low-latency NoSQL database, the summit showcases cutting-edge solutions for extreme-scale engineering.
\nWatch the video.
\nVisit the official Monster Scale Summit website.
\n![\"EloqKV](\"https://www.eloqdata.com/assets/images/monster-5c8e9a4bfa9538c096f4562faaa73fb7.png\")
Redis is a popular in-memory database, but scaling it presents significant issues:
\nThese issues force teams to make trade-offs between cost, performance, and ease of use. EloqKV on ScyllaDB is designed to remove those trade-offs by combining Redis' API with scalability and transaction.
\nEloqKV is a full-featured, distributed ACID transactional database that is compatible with the Redis API. By integrating with ScyllaDB, it delivers:
\nEloqKV follows a multi-threaded, thread-per-core model akin to ScyllaDB, though it is implemented independently without relying on Seastar, ensuring:
\nUnlike Redis, EloqKV supports:
\nEloqKV is more than just a key-value store. At EloqData, we invent a new modular database approach, enabling different compute models to share the same data substrate. This vision extends beyond a single API, with:
\nBy leveraging open-source execution engines, we ensure compatibility with existing ecosystems while avoiding redundant reinvention of wheels.
\nRedis has been a go-to for in-memory caching, but its scaling challenges limit its long-term viability for demanding workloads. EloqKV on ScyllaDB offers an alternative that provides high performance, ACID transactions, and cost-effective scalability—without the trade-offs.
\nIf you’re looking to break free from Redis scaling constraints, EloqKV is the future. Let’s build the next-generation database together and stay connected:\n🔗 GitHub
", "url": "https://www.eloqdata.com/news/2025/03/10/monster-scale-summit", "title": "EloqData on Monster Scale Summit 2025", "summary": "The CTO of EloqData delivered a talk at the Monster Scale Summit 2025, titled \"Overcoming Redis Scaling Bottlenecks with EloqKV on ScyllaDB.\"", "date_modified": "2025-03-10T00:00:00.000Z", "author": { "name": "EloqData" }, "tags": [ "Product" ] }, { "id": "https://www.eloqdata.com/news/2025/02/20/open-source", "content_html": "We are thrilled to announce that EloqData is open-sourcing our core products: EloqKV, EloqSQL, and EloqDoc, as well as the technology underpinning of these product - our Data Substrate component. This marks a significant milestone in our journey to build a more open and collaborative database ecosystem.
\nOpen source has become a cornerstone for database and cloud infrastructures, driving innovation and collaboration across the industry. By providing transparent and accessible code, open source projects like PostgreSQL and MySQL, the Hadoop Big Data Stack, the Spark ecosystem, as well as numerous other projects have revolutionized the way databases are developed, maintained, managed and deployed. These open source projects have not only reduced costs but also fostered a community-driven approach to problem-solving, leading to rapid advancements and widespread adoption. More importantly, these projects inspired a great number of aspiring students, researchers, developers and entrepreneurs to collectively pushing the frontier of data management technologies forward.
\nWe at EloqData have benefited tremendously from open source. We learnt our chops by studying open source projects, developed our product with open source tool chains, and leverage open source projects to enhance and complement the features of our commercial products.
\nThe Data Substrate Architecture is specifically designed to work with open source projects. Data Substrate is a modular, pluggable architecture that can leverage existing open source projects to provide better database solutions without reinventing the wheel. We stand on the shoulders of the giants. By leveraging existing code base, we can significantly reduce the development cost and eliminate API incompatibility.
\nToday, we're taking a major step by making our database solutions available to the open-source community. We are glad to announce that we are open sourcing some of the key components of Data Substrate, to embrace and benefit from the community. Moreover, we are releasing the code base for EloqKV, EloqSQL and EloqDoc – three major products built on top of the Data Substrate technology that implement the Redis API, MySQL API and Mongo API respectively. Included in the source are modifications to existing code bases from other open-source projects to make them work with our Data Substrate. We believe that innovation thrives through collaboration, and we're excited to work with developers worldwide to build better database solutions.
\nA high-performance distributed transactional database with Redis API compatibility. EloqKV offers:
\nA MySQL-compatible distributed SQL database that provides:
\nA MongoDB-compatible document store featuring:
\nAll our products are now available on GitHub. You can find our repositories at:
\nWe invite developers, contributors, and database enthusiasts to:
\nThis is just the beginning. We are committed to:
\nWe believe that by open-sourcing our products, we can accelerate innovation in the database space and help developers build better applications.
\nStay tuned for more updates, and don't forget to star our repositories on GitHub!
\nWe're thrilled to introduce EloqKV, a high performance Redis API-compatible, ACID transactional, scalable, distributed key-value database. You might be thinking, \"Really? Another key-value database?\" In this post, we'll explain what makes EloqKV stand out and the unique values EloqKV offers.
\nEloqKV is the first product built on our groundbreaking Data Substrate technology, an innovative architecture designed to create high-performance, modular, scalable, and transactional databases for the cloud era. We introduced Data Substrate in a previous blog post.
\nLike many key-value stores, EloqKV delivers exceptional performance, handling millions of operations per second on a single server with sub-millisecond latencies. But with its innovative architecture, EloqKV offers unique features that distinguishes it from other key-value databases.
\nMany key-value stores keep all data in memory to achieve low latency. However, this approach significantly increases costs, as even infrequently accessed (cold) data takes up valuable memory. Additionally, durability is often sacrificed for performance, though on modern SSDs optimized Write-Ahead Logs (WAL) can often help reduce the impact to acceptable levels. Moreover, due to the high cost of distributed transactions, most distributed key-value stores either forgo transaction support or offer only limited, partial transactions.
\nEloqKV takes a different approach, embracing the philosophy that ACID (Atomicity, Consistency, Isolation, Durability) transactions are important but should not add extra costs when they’re not needed. EloqKV allows you to enable ACID as a configuration flag. When ACID is disabled, EloqKV incurs no additional overhead, offering performance comparable to that of non-transactional key-value databases. Currently in Beta and available in a future release, EloqKV will allow ACID support to be turned on and off on a per-database basis, further avoiding unnecessary overhead and simplifying management.
\nEven with full ACID, EloqKV incorporates several innovations to minimize overhead. For example, the latency of WAL logging is typically dominated by synchronous writes to stable storage. In EloqKV, logging can scale horizontally and independently, reducing logging latency when more storage devices are added. Moreover, in EloqKV, single-key read transactions do not incur extra overheads. This makes EloqKV blazingly fast for many common workloads. We’ll dive deeper into these technologies in future blog posts or academic papers.
\nEloqKV is designed for scalability from the ground up. Each resource type — memory, CPU cores, logging SSDs, and persistent storage — can be scaled independently. This level of flexibility is super valuable in the cloud era, where resources can be easily provisioned from cloud providers.
\nIf your workload is latency-sensitive and requires all data to be stored in memory, you can reserve virtual machines with large memory capacities. Need to handle a high volume of updates that must be persisted? Simply add a few extra EBS volumes as logging devices. If your data is large, you can use cloud storage options like DynamoDB or S3 to cost-effectively store infrequently accessed data.
\nWe're also developing transparent, dynamic scaling, which will allow you to add and remove resources dynamically as your workload changes. This feature will be available in a public preview soon, so stay tuned.
\nEloqKV is compatible with the Redis API and supports most of Redis's essential data structures. This means that most existing applications can easily take advantage of EloqKV's advanced features with minimal effort.
\nBut EloqKV's ease of use goes beyond API compatibility. The true user-friendliness of EloqKV lies in the fundamental architectural design choices we've made.
\nEloqKV can be deployed as a single-node key-value store, similar to Redis and Memcached. However, EloqKV is inherently designed as a distributed system. It can be deployed in a cluster with ease, instantly benefiting from the high availability and scalability of a true distributed system — no need for a sentinel or special cluster mode.
\nThanks to its built-in cluster awareness and ACID support, an EloqKV cluster behaves almost identically to a single-node server. When connected to one of the nodes, a Redis-compatible client can access and modify data across all nodes in the cluster, potentially with an extra network hop (because physics applies), even if the client is not cluster-aware, and even when the cluster is recovering from failures or changing configuration to add/remove capacity. Additionally, WATCH/MULTI/EXEC commands (i.e., Redis transactions) and Lua scripts can run in the cluster just as they would on a single-node database, eliminating issues like SLOTS and \"CROSSSLOT Keys\" errors.
\nPerformance and cost are always critical factors when evaluating database systems. Numerous database vendors publish articles or white papers claiming performance superiority. It is sufficient to say that we've put a great deal of thought into performance optimizations. We strive to make EloqKV achieve best-in-class performance when features are fairly matched, and we are glad to report that we largely achieved the goal. We encourage you to read our follow up blogs, or try out EloqKV and assess it with your own workloads.
\nOne key point to note is that some of EloqKV's advanced features — such as enabling logging for durability, performing atomic MULTI operations across multiple nodes, and using SSDs to store cold data to reduce memory usage — might be more cost-effective than you expect. Again, we encourage you to test it with your workloads.
Today, we are releasing EloqKV for public preview. This release supports two persistent stores: Apache Cassandra and RocksDB. Cassandra is a disaggregated store that can run on a different set of nodes from EloqKV servers for high availability, while RocksDB is an embedded store. You can learn about how to set up a single node test server or a high-availability cluster following our documents.
\n![\"EloqKV](\"https://www.eloqdata.com/assets/images/eloq_arch_new2-4da3661f1895e2bf97f30463b59919b6.png\")
EloqKV can be deployed in various configurations:
\nAs a cache: When logging and the persistent store are turned off, EloqKV functions as a high-performance in-memory cache, similar to mainstream cache solutions with performance matching the state-of-the-art in-memory cache servers.
\nAs an in-memory database: By enabling logging and the persistent store, EloqKV provides durability while maintaining the same fast read performance as a cache. It flushes writes to the log and applies them to memory-resident data. It also maintains snapshots in stable storage, ensuring full, consistent data recovery after failures or restarts.
\nAs a larger-than-memory database: When the allocated memory is insufficient to store all data, EloqKV evicts data to stable storage and retrieves it as needed. In this mode, EloqKV serves as a strong alternative to many NoSQL databases.
\nAs a full blown transactional distributed database: When deployed in the cloud, EloqKV can be configured as a fully scalable, highly available, and ACID-capable database that can be used as the main store to host primary data. Traditionally this is a postition often taken by expensive RMDBS.
\nWe foresee many scenarios where EloqKV can be valuable due to its unique features and capabilities, and we love to hear from you on your use cases and pain points that EloqKV may be able to address. We are working hard to improve the product and add more capabilities. If you have any questions or suggestions, or want to have an in-person demonstration, please contact us.
\nIn the next few blog posts, we will share some performance benchmarks of EloqKV in different scenarios.
", "url": "https://www.eloqdata.com/news/2024/08/16/eloqkv", "title": "Announcing EloqKV", "summary": "EloqKV is a cutting-edge key-value database offering high performance, scalability, and full ACID compliance. Learn how it revolutionizes data management with its innovative architecture and advanced features.", "date_modified": "2024-08-16T00:00:00.000Z", "author": { "name": "EloqData" }, "tags": [ "Product" ] } ] }