Key Features

Provides precise timing and synchronization with a preemptible real time kernel that achieves lower latencies for advanced application workloads by identifying which specific application threads are high priority processes.

  • High-resolution timers in the real time kernel allows for more fine-grained measurement of time slices.
  • User-specified processes can be assigned the highest priority, ensuring that they will execute without delay, ahead of other system processes. Neither will they be interrupted by other system processes, including things like spinlocks and hard and soft interrupts, which once started, are not interruptible in general purpose operating systems.
  • Optimizes performance with interrupt handling at the device level. With the added flexibility of more fine-grained device level control, system administrators can more easily and effectively reduce system latencies and refine predictability.
  • An adaptive locking algorithm for kernel locks allows tasks to wait in a pre-emptible manner, instead of immediately releasing the CPU when waiting on a contended resource. These adaptive locks effectively decrease the number of context switches on the system, increasing throughput while simultaneously decreasing latency for certain applications.

Increases predictability of critical business process response times through real time scheduler classification and a hierarchical priority scheme to finish on time, every time – even under heavy system loads.

  • In some cases, predictability of response times is even more important than absolute speed and how fast it takes for a process or application thread to execute. Certain process control applications, computer simulations, and high-performance risk management algorithms require the precise synchronization of intermediate execution threads in order for the system as a whole to function properly.
  • The ability to shield or "reserve" key system resources for high priority processes, together with its pre-emptible real time kernel, is what ensures greater predictability of your time-sensitive mission-critical applications. You can be confident that your time-critical applications do not experience resource contention, or unwanted process interrupts that could contribute to increased variability in response and execution times.
  • The real-time scheduler and locking mechanisms help boost performance while maintaining predictability. The spinlocks have been modified to reduce operating system context switching times, dramatically improving the performance of throughput sensitive workloads.

Increases visibility of time-critical applications, enabling you to easily analyze their behavior, identify bottlenecks and improve system performance.

  • In some cases, system latencies cannot be further reduced due to performance bottlenecks that are the result of the way an application was written, or limitations in an application’s architecture. Adding to this challenge, there performance bottlenecks and limiting constraints are difficult to pinpoint precisely, keeping you from maximizing the overall responsiveness of your low-latency systems.
  • Tracing and debugging tools are included that allow you to non-intrusively analyze the run-time behavior of your time-sensitive mission-critical applications. By using these tools you can easily identify resource bottlenecks and target specific areas for improvement.
  • Hardware latency detectors and reporting capabilities improve visibility as well. Latency bottlenecks caused by hardware BIOS (e.g., system management interrupts) can now be more easily identified, while histograms can be used to prioritize latency reduction and optimize performance across the entire hardware and software stack.

Increases efficiency with the ability to mix real time and non-real time workloads in a single virtual machine.

  • Efficient virtualization to separate system components, helping create higher density systems. Both real time sensitive and non-real time sensitive workload virtual machines are supported, co-existing in a single machine and not interfering or affecting each other.
  • Real Time enhancements include the latest stable PREEMPT_RT kernel, support for the latest Precision Time Protocol version and the SCHED-DEADLINE scheduling class. The kernel has inherited hardware enablement to allow customers to run workloads on new hardware platforms. The scheduler predicts task scheduling based on application deadlines, which is particularly useful for real time workloads.
  • Real time applications are more reliable and predictable, so over-provisioned systems can now safely be run hotter, or at higher average utilization, effectively increasing capacity, or alternatively, lowering hardware costs on a per service basis.
  • Virtualization support reduces hardware costs by supporting mixed environments of real-time and virtual machines, or by supporting greater density of real-time instances on a single server.
  • Many trading and market data servers are currently run at 5-20 percent capacity, in order to accommodate trading volumes that can spike up to 15 times their daily average during the course of a day. By using Real Time, these same systems can be provisioned at 10-40 percent average utilization, allowing firms to meet the same quality of service targets and service level agreements, but with half the hardware.

Increases reliability of time-sensitive, business-critical workloads with process and task prioritization and the ability to assign high-priority processes

  • Applications running in a real time environment are more reliable than the same applications running on a general-purpose operating system, especially under heavy system loads or when system demands are both variable and unpredictable.
  • By shielding key hardware resources from extraneous threads and processes, and by assigning only your critical, high priority processes to run on them, you can ensure that your high priority processes will always exhibit predictable performance, regardless of the overall load on the system and the remaining unshielded resources.

Tech Specs

System Requirements

Although SUSE Linux Enterprise Real Time may be run on single processor systems, it delivers maximum benefits on multiprocessor systems. The minimum, and recommended system requirements are as follows:

Minimum System Requirements

512 MB physical RAM

5 GB available disk space

Supported Processors

x86_64

Recommended System Requirements

Multicore / Multiprocessor system

1 GB physical RAM

10 GB available disk space