Welcome!

Java Authors: Liz McMillan, Pat Romanski, Trevor Parsons, Yeshim Deniz, Ivan Antsipau

Related Topics: Cloud Expo, Java, SOA & WOA, PowerBuilder, Virtualization, Security, GovIT

Cloud Expo: Article

Seeding the Cloud: The Future of Data Management

Dramatically improving the agility, flexibility, performance and efficiency of data centers ultimately leads to business success

PowerBuilder Session at Cloud Expo

Doing more with less is a familiar refrain for IT professionals, and today's challenging business environment has only increased the pressure on managers to achieve efficiencies, maximize performance and improve responsiveness of the data center. More and more frequently, IT is turning to virtualization to accomplish its mission-critical goals.

The hot new trend in cloud computing is a natural extension of this drive toward virtualization. In the case of the public cloud, IT can add processing power and infrastructure as needed, and in the case of the private cloud, IT can improve the utilization of existing infrastructure. In other words, cloud computing platforms offer IT the opportunity to increase efficiencies and become more agile, transforming the data center into an environment that delivers greater benefits to end-users.

This tremendous potential of cloud computing can be seen by examining how organizations currently manage, analyze and mobilize data. Cloud has given IT organizations the opportunity to fundamentally shift the way data is created, processed and shared.

In order to take advantage of cloud computing for data management, IT must familiarize itself with the latest issues and trends. This requires a greater understanding of each of the three prominent cloud models - private, public and hybrid - along with proper evaluation of the criteria around a cloud strategy.

Cloud Gazing: Three Prominent Models
Much like those ubiquitous, puffy white masses of water vapor in our stratosphere, each of the prominent cloud computing models has its own unique qualities and benefits. The type of cloud strategy an organization chooses will depend on the specific issues they are addressing.

For IT organizations operating a virtual data center environment, private or internal cloud may offer important advantages. An internal cloud is an effective way to boost productivity while providing users a certain level of self-sufficiency. Such a cloud deployment also mitigates the need to purchase new systems. Through virtualization and cloud infrastructure, users can self-provision a virtual machine on the cloud environment, which can easily be expanded or contracted as a given project reaches fruition. By first developing tools to handle life cycle management of virtual machines, a user can request a virtual machine using a menu to specify resource requirements (CPUs, memory, storage), operating system details, lease period and other miscellaneous configuration information.

An internal cloud can provide critical efficiencies by avoiding data center expansion, faster provisioning of server capacity, faster upgrades of memory, CPU and storage, improved reliability, and business continuity/disaster recovery improvements. Case in point, a major financial services provider recently turned their compute backbone into an internal cloud with a focus on providing a platform to host virtual desktops. The company found that it is easier to share infrastructure with the computational jobs which run after the local trading day ends so there is a nice segregation of "jobs" in the local data center. Additionally, this solution has proven to deliver faster ROI.

Intense interest in public or external cloud strategies is being fueled by the potential for improved efficiencies, flexible and dynamic environments, on-demand infrastructure and smaller maintenance requirements. External cloud services offer the potential for improved business agility, better scalability and versatility. An example of this can be found with one of the world's leading electronic stock exchanges, who turned to an external cloud provider in order to enable brokerage firms to show customers and regulators that best-execution requirements were met for a given trade. An application was written to upload data for every stock in files representing ten-minute increments of trading data based on data from the major exchanges. This allows an accurate reconstruction of the trade environment without having to build out an internal storage infrastructure strictly for emulation (or testing) purposes. A further benefit of this cloud deployment has been the flexibility of pay-as-you-go pricing.

In today's challenging economy, with IT increasingly finding itself strapped for resources, external cloud services are proving to be an effective means for lowering upfront costs and reducing the workload on overburdened IT staff.

Experts agree that the hybrid cloud environment, a mix of in-house and outsourced computing and networking resources, will be a leading choice for enterprises in the near term. The hybrid model offers the greatest flexibility when dealing with the dynamic data requirements of most businesses today.

Regardless of the model, cloud computing is proving to be highly attractive because of its dynamic infrastructure, ability to support any application and operating system, accessibility via Internet protocols, automatic scalability and the fact that no software or hardware installation is required on site. Additional benefits for end-users include the:

  • ability to shift resources from undifferentiated heavy lifting to differentiated value creation,
  • abstraction from infrastructure,
  • "pay for what you use" approach and
  • adaptability to resource requirements

As a result, cloud computing is fast becoming an important tool in the IT arsenal.

Cloud Potential: Classes of Applications Deserving of Deployment
At present, cloud isn't for everyone. In those cases where the most stringent privacy and security policies are in place, a public cloud offering may not meet the regulations needed by IT. Conversely, applications that readily lend themselves to cloud are ones in which the user interface is easily presented in a Web browser and where time sensitivity isn't an issue, such as sales force automation or customer relationship management.

Some classes of applications in use today that might not appear to be obvious cloud candidates will in fact be deployed to the cloud in the near future. Increasingly, we are seeing more functions of data management and data analytics being moved to cloud environments with great success.

One such class of applications is database management systems. For organizations with large databases and high transaction volumes, cloud enables a measure of strategic agility that can meet today's enterprise needs. Sybase Adaptive Server Enterprise (ASE) is a prime example of a mission-critical data management system that allows IT to get a handle on exploding information demands. This application provides access and control over data, transforming information into a vital, accessible and decisive mission asset made all the more effective when utilized in cloud.

As organizations employ larger data warehouses for purposes ranging from standard reporting to strategic business analytics, complex event processing and deep-dive data mining, the need for performance will continue to outpace the capabilities of traditional relational databases. Cloud computing offers IT a powerful means for meeting the performance and scalability requirements of the enterprise data warehouse. With large-scale systems continuing to expand, alternate approaches to support standard reporting, analysis, and power-user ad hoc queries will become increasingly established as the platforms of choice for very large database systems.

As with other cloud environments, analytics in the cloud benefit end-users by offering a pay-as-you-go model and adaptable resource requirements that free up IT from the need to purchase additional hardware and going through the extensive procurement process.

In order to satisfy the rapidly expanding need for analytical performance, Sybase offers a solution well suited to exploit the benefits of cloud. Sybase IQ takes an alternate database approach, storing data oriented by columns instead of rows.  This approach has proven superior in sustaining the performance and rapid growth requirements of analytical applications and, when combined with cloud computing, offers significant advantages. The column-oriented methodology provides a combination of architectural simplicity and the ability to configure data in a way that can reduce the physical amount of data that must be accessed. Reducing the storage footprint while optimizing column access will reduce data access latency and improve use of network bandwidth, thereby contributing to a scalable environment that continues to provide consistently high performance as data volumes, number of users and number of queries increase.

With a growing trend toward mobility in the workforce, IT faces increased demand for access to mobilized applications. Managing data and exchange technologies can place significant demands on IT, necessitating regular onsite maintenance of mobile devices. Cloud offers IT the scalability, flexibility, performance and responsiveness needed to deploy and manage mobile applications. For example, Sybase SQL Anywhere provides data management and data exchange technologies that enable the rapid development and deployment of database-powered applications. Design and management tools within SQL Anywhere enable IT to implement and deploy mobile applications, while also easily providing support.

For one major telecommunications provider, cloud computing is key in offering Web-interfaced applications hosted and deployed in a managed service (SaaS) model. This company leverages Sybase's Afaria solution, which provides comprehensive management and security capabilities to ensure that mobile data and devices are up-to-date, reliable and secure.  The provisioning and scalability advantages of cloud make it possible for the company to offer an integrated suite of managed mobility applications for their customers that can be contracted individually or in combination. Afaria enables the company to offer its hosted applications, and assist customers in managing the deployment, expenses and ongoing support to a global mobile workforce without technical complexity and security risks.

Cloud computing can enable IT to more effectively handle the wide ranging database requirements of mobile workers, while minimizing the impact on end-users, thus allowing them to focus on the work at hand rather than getting bogged down with the technology.

Multi-tenancy Requirements
There are numerous considerations IT should take into account when pursing a data center cloud strategy. These range from security and compliance requirements to the need for a rich user interface, effective management and control, and the need to establish standards, provide support and improve performance.

In the case of multi-tenant applications, IT must consider the separation of data from a security, privacy, and compliance perspective. Data that is bound by strict privacy regulations, such as medical information covered by the Health Insurance Portability and Accountability Act (HIPAA), will require that tenants log in, after which they are then routed directly to their own secure database server.  In this way, data remains separate and secure, and in compliance with government privacy mandates.

Another factor IT must consider is the implication of application performance on end users. If one tenant is running a large, complex report, overall performance on the database server can be slowed. IT must gauge the sensitivity of other users to such performance issues and factor this into decisions regarding external cloud providers as well as internal cloud initiatives

In terms of choosing cloud schema, IT can establish a single database on the server with separate schemas, allowing one tenant to employ a set of tables within the database while another tenant is able to utilize a different set of tables within that same database. In this scenario, the cost of provisioning is lower, but there is a trade-off in performance and data isolation. A second option is to employ shared schema where all tenants' data is contained in the same table. Each tenant has a column in the table, making this option more cost effective to provision, however application development becomes more complex.

Operational Considerations of Cloud
When hosting data in the cloud, IT must take various operational considerations into account. These considerations should include how backups will be performed and how often data will be backed up. Additional operational details, such as the existence of offsite storage and the robustness of disaster recovery protocols should be examined.

A further consideration IT should take into account as cloud computing gains wider acceptance is the ability for users to access the data center from both desktop and mobile environments. When considering mobile access, IT must evaluate technical considerations, such as how users will traverse an organization's firewall. Existing security requirements around data, and the ability to synchronize that data to mobile devices when appropriate, clearly add to the complexity of cloud deployment. Areas such as these should be top of mind for IT as they move forward.

The Sky Is the Limit With Cloud
As IT seeks to maximize the efficiency, performance and responsiveness of the data center, virtualization and cloud computing are providing important new tools to meet mission-critical goals. Cloud is opening the door for IT to get the most out of vital processing power and infrastructure.

The tremendous potential of cloud computing is fundamentally shifting the way data is created, processed and shared. Currently, the most prominent cloud models, private, public and hybrid, are transforming the data center into an environment that delivers significant benefits to end-users, in how they access and manage data.

As new strategies evolve and more classes of applications become viable in the cloud, IT has the opportunity to dramatically improve the agility, flexibility, performance and efficiency of data centers which, in turn, will ultimately lead to bottom-line organizational success.

More Stories By Irfan Khan

Irfan Khan is CTO of Sybase. He oversees all the technology offices in each of Sybase's business units. Together with architects residing within the technology offices, Khan ensures the customers' voice and needs of the market are reflected within the company's innovation and product development. In addition, he oversees Sybase technology evangelism efforts.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.