Software-defined computing

Struggling with costly grid infrastructure silos? Here’s a solution.

Share this post:

Your organization might employ a cluster or grid management solution to support its high-performance distributed computing environment. But the solution requires a separate infrastructure to support each application workload. How do you preserve line of business ownership while delivering service-level guarantees? Better yet, how do you address ongoing pressures to reduce infrastructure costs and meet your organization’s demand for higher-quality answers fast?

IBM® Platform™ Symphony V7.1 is helping address these challenges with several new capabilities, including its latest add-on product—IBM Platform Application Service Controller for Platform Symphony. The resource-sharing model for Platform Symphony software makes it practical to deploy multiple heterogeneous applications on the same shared grid. By sharing resources fluidly while preserving ownership, resources are utilized more fully, delivering better performance and helping reduce infrastructure costs for your organization.

In the past, if it was possible for frameworks to coexist on the same cluster at all, it was necessary to partition resources statically. The problem with this approach is that each framework essentially exists in its own “silo.” So if one application is idle, other applications cannot take advantage of idle capacity—resulting in resource inefficiencies and suboptimal performance.

Distributed computing on a single, shared infrastructure

With IBM Platform Symphony, you can support a broad set of distributed frameworks including multiple instances and versions of the same framework concurrently. By consolidating applications and running them on a single shared resource pool, individual applications can flex-up and harness resources that would normally be dedicated to other applications—significantly improving performance and increasing the overall efficiency of the environment.

Platform Application Service Controller is able to leverage Platform Symphony’s sophisticated resource-sharing model, and extends Platform Symphony’s capabilities to a broader set of application services than ever before. Platform Symphony V7.1 provides granular resource sharing among a wide range of compute-intensive and data-intensive workload types, including long-running services.  It supports batch, high-performance service-oriented architecture (SOA), Hadoop MapReduce and a variety of other “born-on-the-cloud” service frameworks.

Overcoming management woes

Another challenge with distributed computing clusters is that they can be difficult to manage. Organizations are forced to deploy separate environments, each with their own unique management, monitoring and reporting interfaces. Platform Symphony provides a unified management interface that allows administrators to manage multiple distributed frameworks. Administrators can see where application services are running, monitor the status of services, and see how resources are being allocated.

With monitoring, reporting, usage accounting and capacity planning information now consolidated, administrators can avoid the need to collect and aggregate metrics from each framework individually. Resources can be more efficiently monitored to meet service-level objectives, and new resources can be quickly brought online as needed. You can optimize existing hardware resources and defer the need for incremental capital investment. Bottom line: you not only can reduce costs by using infrastructure more efficiently, you can reduce management complexity as well.

For more information on IBM Platform Symphony and IBM Platform Application Controller, visit: ibm.com/systems/platformcomputing/products/symphony

More stories

Introducing Spectrum Storage for AI with NVIDIA DGX for deep learning

AI, Cloud object storage, Deep learning...

Artificial intelligence, machine learning and deep learning projects are happening in every industry and research team across the globe. According to IDC[1], by 2019 40 percent of digital transformation initiatives will use AI services; by 2021, 75 percent of commercial enterprise apps will use AI, over 90 percent of consumers will interact with customer support ...read more


Guard your virtual environment with IBM Spectrum Protect Plus

Banking & finance, Cloud object storage, Data centers...

The server virtualization market is expected to grow to approximately 8 billion dollars by 2023. This would mean a compound annual growth rate (CAGR) of 7 percent between 2017 and 2023. Protection of these virtual environments is therefore a significant and critical component of an organization’s data protection strategy. This growth in virtual environments necessitates ...read more


Spectrum Scale, SDI and motorcycles: Scale up, scale out!

IBM Systems Lab Services, Services, Software...

You might wonder if I’ve completely gone off the rails when I mention my passion for motorcycle riding in the same breath as Spectrum Scale and software-defined infrastructure (SDI). Trust me, I’ll make a case for why I put them together. In the early days, a motorcycle was nothing more than a bicycle with a ...read more