The In-memory Database System (IMDS) as a Powerful, Persistent Memory Caching Solution
Download the white paper, “NoSQL, Object Caching & IMDSs: Alternatives For Highly Scalable Data Management“.
Has the RDBMS reached its limits in today’s highly scalable Web-based applications?
While object caching (such as Memcached) and NoSQL solutions may fit some scalable, high performance systems, many high volume applications still benefit greatly from “real” database features that support persistence, recoverability, ease of programming and data design, data integrity and more.
The 64-bit in-memory database system (IMDS) combines benefits of both categories: the fastest possible Web data access, along with the reliability, recoverability and programming/design features of a proven database system.
For example, McObject’s eXtremeDB-64 is increasingly popular as a caching layer for financial, Web-based social networking, SaaS and other emerging applications.
The Problem
Business and social transactions are rapidly moving to the Internet. Data sets supporting Web-based financial, commercial, e-commerce, and social networks are growing fast, but as they scale, quality of experience can vary widely between visits as popular sites buckle under the load of heavy traffic. Too often, the information needed to satisfy a user, or to close a sale, is trapped in a back-end DBMS.
Disk-based storage severely cramps data availability. Traditional object caching solutions provide speed by keeping some records in memory. But they lack means for recovery or ensuring data consistency in the event of system failure. NoSQL solutions are scalable and address specific problem domains, but they, too, lack features such as ACID transactions, formal data access and definition languages, and more.
The Solution: eXtremeDB-64
The eXtremeDB-64 solution blends the highest speed and data availability with the durability, safeguards on data integrity, and sophisticated querying capabilities that are DBMS hallmarks.
Positioning eXtremeDB-64 as the front-end for a traditional database lends high volume systems the benefits of all-in-memory data access. It also ensures continuous operation and data consistency, proven in a wide variety of applications.
eXtremeDB-64 features include:
- Transaction logging
- A Cluster module that multiplies net available processing power, ensures availability/uptime and enables cost-effective system expansion
- A High Availability module for master/slave replication with automatic failover
- 64-bit support for very large databases, proven at 1+ terabytes, 15 billion rows and beyond
- Pure in-memory design eliminates I/O, cache management, data transfer and other overhead, resulting in the fastest possible responsiveness
- Progressive error-detection and other sophisticated debugging features
- Multiple APIs: native/navigational C/C++, SQL ODBC & JDBC, and Java Native Interface. C# can be used with the ODBC .NET Managed Provider
- eXtremeDB Data Relay for efficient, fine-grained data-sharing with external systems
Resources
- eXtremeDB and eXtremeDB-64 product information
- Download the white paper, “NoSQL, Object Caching & IMDSs: Alternatives For Highly Scalable Data Management“
- Watch the recorded Webinar, “Data Acceleration and Business Continuity With McObject’s eXtremeDB Cache“
- White Paper: Scaling eXtremeDB-64 Beyond the Terabyte Size Boundary
- White Paper: What Makes A Database System ‘In-Memory’? In-Memory Database Systems: Myths and Facts
- Read about eXtremeDB for HPC independently audited benchmark test results at STAC. Learn more about the hybrid in-memory and/or persistent database fast enough for financial systems.
Proven
Find eXtremeDB in over 30,000,000 deployments worldwide.
We’re honored to be used world-wide by innovative companies that lead their industries. See a partial client list.
Originally designed as an in-memory database system.
Learn why this matters.
Embedded & Real-time Systems
The eXtremeDB in-memory database system was designed specifically for use in resource-constrained, mission-critical and safety-critical embedded systems.
Learn more about eXtremeDB in-memory database for embedded systems
Learn what makes eXtremeDB ideal in real-time systems
High Performance Computing
eXtremeDB in-database analytics offers breakthrough efficiency and can be used with the product’s in-memory database system (IMDS) capability, or independent of it.
Learn about hybrid eXtremeDB for Big Data and Analytics or for Financial systems
Learn more about the IMDS as a powerful, persistent memory caching solution.
Hybrid In-memory and/or Persistent
Combine both database paradigms – in-memory and on-disk – in a single database instance. Specifying one set of data as transient (managed in memory), while choosing persistent storage for other record types, requires a simple database schema declaration.
Learn why starting with an in-memory database makes for a better hybrid system.