Continuous Application Shadowing™
One of the major recent trends in storage is the availability of low-cost bulk disk storage. This enables the commoditization of replication, which in the past has been a technology that has been affordable only for the largest enterprises.
Replication has to be application aware in order to maintain complete integrity and consistency of the application and recover from different types and combinations of failures. Block level replication and continuous data protection approaches, while suitable for flat files, simply do not work for today's complex applications.
Mimosa provides Continuous Application Shadowing™ using low-cost disk storage and application aware replication technology based on Transaction Log Shipping. The NearPoint server continuously shadows multiple live production servers. A copy of the application data on the NearPoint server is continuously updated as the production server data changes. This continuous update process is non-disruptive to production servers – the production servers continue to operate at full performance.
In the Mimosa NearPoint architecture, commodity replication & application awareness enables Continuous Application Shadowing, which is the basic mechanism for enabling recovery, archiving and advanced compliance functions.
Smart Extraction
Application shadowing is an efficient bulk approach to get data to the NearPoint server. The NearPoint server subsequently parses the application data and separates all the data objects in a process called Smart Extraction. This enables retrieval of application objects and items at any level of granularity. For example, with Microsoft Exchange, the Smart Extract process separates all messages, attachments, contacts, calendars, etc. for classification upon receipt of bulk data from the Exchange servers. As a result of Smart Extract, retrieval can be done not just at the Exchange database level but also at the mailbox or message level – all with a single pass over the production Exchange data.
Indexed Object Repository
After the parsing of data objects, additional metadata is associated with the data objects. This is done to make the search of data more flexible and powerful. It is also to enable new data management capabilities. Examples of additional metadata are full-text index information for search and Digital Signatures for tamper protection.
The Indexed Object Repository is a database on the NearPoint Server that holds all the metadata associated with data objects. The repository captures information across time and space. Business users and compliance managers can access the Indexed Object Repository to quickly search, discover and retrieve information in a powerful and flexible manner. The repository also manages all retention and disposition rules as specified by the organization to comply with business record keeping requirements. Single instancing and compression reduce duplicate objects and reduce object size to significantly reduce the storage footprint.
New Generation of Data Management Applications
The Indexed Object Repository provides a common application framework for a new generation of data management applications such as recovery, archiving, advanced discovery & search, monitoring & supervision and disaster recovery that require access to application data across time and space. These data management functions will be provided by Mimosa and third parties. With the NearPoint architecture, data management is offloaded from production servers resulting in improved performance and robustness of production applications.
Mimosa NearPoint Architecture Advantages
The Mimosa NearPoint architectural approach has a number of significant advantages:
- Unified Data Management - Mimosa NearPoint enables integrated archiving and recovery that share a common underlying framework and the indexed object repository. The key advantages of unified data management are consistent retention & disposition management, reduced cost of storage (no stove-piped silos!) and a single management console.
- One Pass Protection™ – With Mimosa's One Pass Protection, application data is captured efficiently in bulk at higher level granularity but restore and retrieval of application data is at any required granularity. One Pass Protection is the most scalable way to capture application data and enable retrieval at different granularity levels. As an example, with NearPoint for Microsoft Exchange, immediate restore can be done at the database, mailbox or message level. This is in contrast to traditional tape backup approaches where multiple time-consuming backup passes (“brick backup”) are required to enable restore of mailboxes or messages. This approach is also in contrast to traditional email archiving, where data is copied message by message from production Exchange servers using MAPI resulting in a heavy load on the Exchange server and a large archive window.
- Application-Intelligent Data Capture – Mimosa's Application Shadowing is at the transaction log level ensuring application integrity and consistency. In addition when transaction logs are processed at the NearPoint server if there is Exchange data corruption it will be detected immediately. This is in contrast to traditional snapshot and CDP approaches where data is captured at the block level resulting in inconsistent copies of the Exchange database.
- Scalability – The Mimosa NearPoint architecture has been built from the ground up for scalability and has significant advantages over the first generation email archiving solutions. The very high performance and low impact capture of data from Exchange at the database level is in contrast to the slow and high impact “MAPI” based approaches of extracting data from Exchange. The “handler” based Mimosa architecture is designed to allow different NearPoint modules to be distributed across different servers for the highest level of scalability.
- Chain of Handling of Messages – Mimosa collects data from Exchange at the transaction level which allows the entire chain of handing of messages to be captured. It is critical during investigations to reconstruct the different transitions of a message in an individual's mailbox – the movement to different folders, and through modification and deletion. Furthermore the Mimosa approach allows the capture not only of email but also all calendar items, contacts, notes, etc. throughout their lifecycle.
- Co-existence with Tape Backup – Mimosa NearPoint co-exists with tape backup solutions from vendors such as Veritas, Computer Associates, IBM and EMC Legato. Tape backup is important for offsite transport and deep archiving. With NearPoint, production servers can still be backed up with tape backup agents on production servers. Production servers can alternatively also be backed using tape backup of the near real-time application shadow copy on the NearPoint server, thus eliminating the backup window problem.
- Zero Footprint Deployment – One of the most important advantages of the Mimosa NearPoint architecture is the ease of deployment. There are no agents on the production server assuring worry free deployment with no danger of causing performance and availability problems in production environments. There are also no agents on desktops enabling rapid enterprise-wide deployment across tens of thousands of desktops. All processing of messages is done off-host imposing no load on production systems. Mimosa NearPoint is designed to be installed as a packaged solution and it has built-in policies that enable the installation and deployment in a few hours.
|
