Resources
Get Started with MemFlow
Quick and easy setup in minutes
Download and install the MMS (MemFlow Management Server) on any Linux machine, then connect to the web console from any device. Connect your PostgreSQL or Oracle database and start caching hot tables immediately.
Schedule a Live Demo
Book a 15-minute session with our support team. We'll show you around MemFlow and answer all your questions.
Schedule NowDemos & Documentation
Geting Started
Introduction to MemFlow and basic concepts
Installation
Step-by-step installation guide for MMS and MDS
Create a MDS Cluster
Setting up and configuring a distributed MDS cluster
Adding Tables to Cache
How to select and cache tables from your database
Querying Cache Tables
Accessing and querying in-memory cached data
Create Native Cached Tables
Creating tables directly in MDS without a source database
Auditing
Monitoring and auditing database activities
Performance & Resource Tuning
Monitoring and auditing database activities
Frequently Asked Questions
In a sentence, what does MemFlow do?
MemFlow accelerates PostgreSQL and Oracle databases by transparently caching frequently-accessed tables in distributed in-memory storage, delivering up to 100x faster query performance without requiring any application code changes or schema modifications.
What is a hot table?
A hot table is a database table that receives frequent read queries and high traffic, making it a prime candidate for in-memory caching to significantly improve application performance and reduce database load.
Why don't I just use a bigger postgres cache or pin tables into Oracle memory?
While pinning tables to the database's buffer cache helps, it has critical limitations: the cache is shared among all tables competing for limited space, performance degrades when cache memory fills up and causes evictions, and most importantly, it doesn't scale beyond a single server—whereas MemFlow provides dedicated distributed memory that scales horizontally across multiple nodes, guarantees your hot tables stay in memory without eviction pressure, and can grow independently of your database server's constraints.
Why don't I use a dedicated in-memory database?
Using a dedicated in-memory database requires significant effort: you must rewrite application code to query a different database, manage complex data synchronization between your primary database and the in-memory system, handle schema changes in two places, and maintain separate failover and consistency mechanisms—whereas MemFlow works transparently as a caching layer that requires zero code changes, automatically syncs data, and lets you keep your existing database as the single source of truth with minimal operational overhead.
How does MemFlow scale?
MemFlow scales horizontally by distributing tables across multiple memory nodes using consistent hashing—you can add more nodes to increase total memory capacity and query throughput, with each node handling a portion of your cached tables, ensuring that as your data grows or traffic increases, you simply add more nodes to the cluster without any downtime or application changes.
Will I pay a higher Oracle license?
No—MemFlow doesn't increase your Oracle license costs because it operates as an external caching layer using standard database connections (JDBC/ODBC), so Oracle only sees it as a regular client application, not additional database instances or cores that would trigger higher licensing fees; in fact, MemFlow can save you significant money because without it, you'd likely need to upgrade to more powerful Oracle servers with additional CPUs to handle performance demands, and since Oracle licenses are priced per processor core (often $47,500+ per core plus annual support), avoiding even a single server upgrade can save hundreds of thousands of dollars.
I'm stuck with an old Oracle version. How can MemFlow help?
MemFlow is the perfect solution for your situation—it gives you modern in-memory performance acceleration without requiring any Oracle upgrade, so even if you're stuck on Oracle 11g and missing features like Oracle 12c's In-Memory Column Store or later versions' performance enhancements, MemFlow provides a separate in-memory layer that works with any Oracle version (even 9i and 10g), delivering dramatic speedups for your hot tables while letting you stay on your current licensed version and avoid costly upgrade fees or re-licensing negotiations.
Can MemFlow help with ETL workflows?
Yes—MemFlow's "ETL killer" feature eliminates most traditional ETL complexity by making any cached table from any registered database (Oracle or PostgreSQL) instantly accessible to all other connected databases at in-memory speeds, so instead of building complex ETL pipelines that extract, transform, and load data with hours of latency, your applications can directly query live data from any source database as if it were local, with sub-millisecond response times and no data staleness, effectively replacing batch ETL workflows with real-time data sharing across your entire database ecosystem.
What is a MemFlow cluster and how does it help my business?
A MemFlow cluster is a group of two or more MDS (MemFlow Data Servers) that work together as a unified in-memory caching layer—if one MDS server fails, the cluster automatically redistributes and re-caches its tables across the remaining servers, ensuring zero downtime and continuous high-performance access to your data; this high-availability architecture protects your business from service interruptions, eliminates single points of failure, and provides the reliability and resilience that mission-critical applications demand, all while maintaining the same blazing-fast in-memory performance across your entire database infrastructure.
How does MemFlow compare to Redis?
While Redis requires you to rewrite application code to use a key-value API and manually manage what data to cache, MemFlow works transparently with zero code changes—you simply point your database connection to MemFlow and it automatically accelerates your existing SQL queries. Redis is excellent for session storage and microservices, but MemFlow is purpose-built for accelerating traditional relational databases (PostgreSQL and Oracle) with automatic cache management, cross-database access, and enterprise features like replication sync and audit logging that Redis doesn't provide out of the box.
Do I need to change my application code?
No—MemFlow requires absolutely zero code changes to your applications. Simply update your database connection string to point to MemFlow instead of directly to your database, and MemFlow transparently intercepts and accelerates your SQL queries. Your applications continue using the same SQL statements, database drivers, and ORMs they've always used, making deployment as simple as updating a configuration file.
What happens to my data if MemFlow goes down?
Your data is completely safe—MemFlow is a caching layer, not your primary data store, so all your data remains intact in your original PostgreSQL or Oracle database. If MemFlow becomes unavailable, queries automatically fail back to the source database (with Entry edition) or instantly redirect to other healthy nodes in the cluster (with Standard/Enterprise editions). There is zero risk of data loss, and in high-availability configurations, users may not even notice a brief performance dip during failover.
How do I migrate from Entry to Standard edition?
Upgrading is seamless—simply purchase a Standard license and import the new license key into the MMS web console. There's no reinstallation, no downtime, and no data migration required. The moment you activate the Standard license, advanced features like automatic mode, multiple replicas, and real-time replication immediately become available. You can test Standard features during your free trial period to experience the full capabilities before committing.
Is there a minimum contract period?
No—MemFlow operates on a simple monthly subscription with no long-term contracts or commitments. You pay only for the memory capacity you use each month, and you can scale up, scale down, or cancel anytime. The first month is completely free, giving you 30 days to test MemFlow risk-free before any charges begin. This flexibility means you're never locked in and can adjust your deployment as your needs change.
Can I select all my hot tables and let MemFlow handle memory management?
Yes—absolutely! You can select all your hot tables and let the MemFlow cluster intelligently manage them; it will automatically cache as many tables as possible across your available MDS memory, prioritizing them based on your configuration, and when memory is full, it maintains a queue of remaining tables so that as soon as you remove a cached table or add more MDS capacity, the next table in line automatically gets cached without any manual intervention, giving you a worry-free "set it and forget it" experience that maximizes your available memory resources.
Can MemFlow operate in automatic mode for non-experts?
Yes—MemFlow is designed for users who aren't database experts! It operates in three modes: manual mode where you select specific tables if you know what you need, semi-automatic mode where MemFlow shows you I/O statistics and recommendations next to each table name so you can make informed decisions, or fully automatic mode where MemFlow's intelligent table loader analyzes your database activity daily, identifies the hottest tables based on real-world usage patterns, and automatically caches them without any intervention—so you can simply turn it on and let MemFlow optimize your database performance while you focus on your business.