– Louise Funke, Vice President of Marketing at TransLattice (www.translattice.com), says:
Despite significant investments directed at avoiding downtime, businesses continue to suffer significant outages. Often, these outages are due to factors beyond their control and are fundamentally due to the complexity and fragility of the typical application infrastructure stack. The TransLattice Application Platform (TAP) provides a new way of delivering business-critical applications that eliminates any single point of failure and dramatically reduces the complexity of implementing and scaling applications.
TAP weaves together all of the disparate components and middleware that are in the data center: the storage, the database, the caching, the load-balancing, the optimization to create a single, tightly-integrated node. These nodes are mass market servers that are rack mountable. When more capacity is required, you simply add a node. Nodes can be geographically distributed anywhere on a network. There is no master node. All nodes are equal and the cluster can be managed remotely from anywhere. With this architecture, computing resources can be placed closer to where the data is needed by users or clients. Because of the distributed architecture, if one of the nodes does happen to go down, the system itself doesn’t go down. It continues to operate because there is enough redundancy built into all aspects of the system.
TransLattice nodes can be either physical appliances or cloud instances and a cluster can be comprised of a combination of physical and cloud instances. TransLattice provides unique policy-driven data management capabilities that allow public and private clouds to co-exist efficiently, ensuring that sensitive data remains on private nodes.
The world’s first truly globally distributed relational SQL database spans all nodes in the cluster and presents one massive, coherent database to users and developers.
Google and Facebook have created ways to spread search, indexing and other workloads across many, many identical computers, and thereby eliminate the impact of discrete failures. This has been extremely expensive in terms of building large server farms and engineering resources, but it is worth it due to the scale of its operations. The reduced operational costs due to high-availability and reliability justify the added development costs these behemoths have incurred.
TransLattice provides similar types of benefits to all organizations for business-critical transactional applications such as ERP, CRM and SCM, while maintaining the existing development model and without re-writing applications.
TAP provides a truly scalable and elastic cloud infrastructure to improve access to applications and data, particularly for global business, with a physical appliance fueled by a geographically distributed relational SQL database management system that moves data closer to where it is likely to be used. The built-in redundancy eliminates any single point of failure making traditional DR solutions obsolete.
Where should the TransLattice technology rank in terms of overall priority in the data center?
TransLattice should be the top priority for companies looking to eliminate unexpected outages of business critical applications. The current technology is unsustainable. Application stacks are comprised of numerous components and interfaces that create a fragile infrastructure subject to various types of failures. It is extremely difficult to anticipate all the many components and ways in which the infrastructure might fail.
TransLattice collapses the application stack into a single, tightly-integrated and easily managed appliance or cloud instance called the TransLattice Application Platform or TAP. Applications are installed in an application container TAP and then deployed as nodes in a cluster. Each node, whether a physical appliance or cloud instance, is identical and nodes cooperate to deliver the application in a scalable and redundant fashion. If a node fails, traffic is automatically redirected to other nodes without disruption of service.
The biggest challenges for data center and IT managers:
The biggest challenge TransLattice helps data center and IT managers overcome is the need to make business process conform to IT constraints. TransLattice can dramatically change how IT and the line of business operates. Today, many organizations adapt their business according to restrictions of the IT infrastructure. For example, a major financial institution maintains five installations of their banking infrastructure in different regions for regulatory compliance. They are unable to offer certain services in various regions so handling the needs of multi-national customers can be difficult. It also means that they do not have a minute-by-minute picture of their entire business and where money is at any given moment. A US-based software as a service company has all of its infrastructure in the US. Because of regulatory compliance issues they must recognize all revenue here in the United Sates. They are unable to take advantage of more favorable tax jurisdictions for transactions that actually occur outside the US.
How can data center and IT managers overcome those challenges?
TransLattice is a new approach to transactional computing that helps IT adapt to business rather than the other way around. It also provides a more resilient IT infrastructure that can improve business continuity.
TransLattice has a policy definition framework that allows users to define redundancy policies that control the minimum number of copies of data retained. For example, an organization may define that a minimum of three copies are stored on at least two different nodes on two continents. This would ensure that should a disaster affect an entire continent, data would still be available and operations could continue uninterrupted. Users can also specify data locating policies that control such things as ensuring that certain types of data not be stored in cloud instances or must remain on nodes in a certain country to meet compliance requirements.
TransLattice keeps track of historical access patterns such as what portions of tables are accessed together from what locations. It determines, based on this information, combined with defined policies and understanding of the cost of certain network links, the lowest cost, most efficient access and storage plans that optimize user experience and optimize use of available resources. This allows all of the redundancy to be done at the database layer rather than at some other underlying mechanism such as the Storage Area Network and for the redundancy to be aware of how data is actually used and position it closer where it is ultimately going to be consumed so queries can be run in a highly efficient manner. It also provides linear scaling as additional nodes are added to the TransLattice cluster. The transactional throughput increases linearly.
The TransLattice patent-pending globally distributed database and consensus algorithm provides a consistent view of the database across all nodes. All nodes look and act like a single database even though the data is actually shared across various nodes of the system. The TransLattice database is an ANSI-SQL compliant, relational database management system that supports ACID-semantic transactions and all updates and changes can be made in real-time. Its consensus protocol is integrated with a local write-ahead log mechanism to ensure that transactions that require durability obtain it before the commitment completes. The consensus protocol regulates all changes to the database. Parallel propagation of changes is inherent to TransLattice’s management process, so change is orderly and controlled. The database understands data usage patterns and use cases and automatically moves data where it is needed while ensuring that data jurisdiction requirements are adhered to. TransLattice is able to adapt to changes in usage patterns and in the environment to use all resources most effectively.
TransLattice has worked to eliminate technical risks by making the transition to TAP as simple as possible. Because its distributed database is ANSI-standard SQL, built on a relational model, enterprise Java applications should run without modification. This also means that IT managers can move a single application to the TAP platform and existing interfaces will continue to work.
When IT managers take a close look at other technologies, they will see that TransLattice is unique. Most distributed solutions fall into three categories: locally distributed; NOSQL; and mem-cache solutions. Each of these have significant shortfalls. Only TransLattice anticipates workers’ application and data needs based on organizational policy and past usage, delivering the information when and where it is needed without having to modify applications or risk data loss. By positioning data closer to end users and adhering to compliance policy proactively, TransLattice supports relevant business processes while continuing to leverage the existing investment in enterprise SQL-based applications.
TransLattice provides a single, unified mechanism that delivers redundancy; places data where it is ultimately going to be used, in compliance with policy, and ensures the database remains a high-performance platform for business-critical transactional applications. TransLattice provides improved resilience, a vastly easier way to scale enterprise applications, improved performance for remote users and an easier way to move enterprise applications to the cloud.
