Linux reliability and availability

Although Linux is widely acknowledged as being extremely robust and reliable, work is proceeding at a furious pace to enhance those features even more. In the area of the disk file system, these features increase system reliability and enable recovery in the event of an unexpected failure:

Journaling

file systems for Linux eliminate the need for a file system consistency check after an unintended shutdown due to a power outage or other similar event. A Logical Volume Manager (LVM) included with Linux lets the user dynamically resize disk partitions on the fly, as well as set up software-based RAID systems.

Additionally, there are features that you can use for greater improvements in mission-critical applications requiring high reliability and availability. Such out-of-the-box benefits include:

Reduced reboots – adding new software should not require your customers to reboot their systems at all. In addition, changing items like IP addresses, protocols, and device drivers, requiring rebooting on some operating systems, does not on Linux. Automatic system recovery and logging tools for total system management. Network servers for truly distributed environments Loadable device drivers for added configuration flexibility and system availability

Although most out-of-the-box features will be available simply by migrating your solution to Linux, even more reliable solutions can be obtained by making the most of the services via new hardware configurations. These features include:

RAID support for resilient data storage Standards-based redundant network links for improved network connectivity and reliability Redundant power supplies with power monitoring services for improved resilience to environmental factors High-availability server chassis with hot-swap capability for increased solution availability, thanks to improved fault isolation and recovery.

There are also a number of underlying capabilities that can be made the most of by the Dialogic system software and/or the customer solution:

  • LDAP (Lightweight Directory Access Protocol);
  • CORBA* (Distributed Object Broker);
  • 64-bit processor support for Itanium processors;
  • Enhanced message Internationalization and Localization;
  • X Windows Graphical User Interface (GUI);
  • Dynamically Loadable Device Drivers;
  • Symmetric multiprocessor (SMP) support;
  • Apache Web Server;
  • Plug-N-Play device;
  • Java* Virtual Machine; and
  • Clustered system support.

Enhanced Performance At Lowered Costs

For those solution providers who aim to increase throughput while simultaneously lowering costs, Linux offers considerable benefit. For instance, Linux support for I2O (a dedicated I/O processor) bandwidth-intensive applications allows certain applications to offload their I/O load to dedicated processors, speeding up the overall system.

Hardware requirements are no greater for Linux than they are for Windows NT 4.0 or Windows 2000, and you don’t usually need to upgrade your hardware to get the benefit of this OS.

Linux is license free, so developers, integrators, and end-users do not need to purchase costly “client-access” licenses. A small portion of those savings invested in hardware upgrades will typically yield a significant benefit to everyone involved.

Additional benefits can be realized by proactively making the most of the following features offered by the OS:

The top tool for processor accounting and RAM allocation scatter/gather I/O for streamlining moving data from non-contiguous RAM to contiguous disk Processor affinity to allow threads to be locked to a particular processor where cache hits are more likely to improve performance

More flexible/proc interface for system tuning

Enhanced Scalability for Incremental Growth

Scalability allows systems to grow when expansion is needed. This feature is particularly useful for companies that start small due to financial limitations, or in a business model in which capacity growth is tied to subscriber or revenue growth. A scalable solution lets these customers install a small system and increase capacity as needed without having to perform major changes or what is sometimes called a “forklift upgrade” on their base machine.

Enhanced scalability in Linux is provided out-of-the-box via Symmetric Multiprocessor (SMP) support and a common API across the various Linux variants. Even the lowest-level variant of Linux has SMP support. As a result, even desktop machines can make the most of the benefits of having multiple processors in a single machine. Scalability in Linux permits steady growth of up to 16 processors within a single server.

Linux implements a common API across many different types of hardware. The same application can therefore run unchanged on a small laptop computer as well as on a large multiprocessor back-end server. Furthermore, the same Linux API is supported on an embedded configuration (one with a minimal memory utilization and without a monitor, keyboard, or mouse – sometimes referred to as “headless”) so you can port your solutions to embedded systems with few application changes.

Other out-of-the-box scalability benefits of Linux include:

Dynamic SMP allocation for applications lets developers and system administrators reconfigure their systems to optimize for changes in SMP hardware and SMP needs, even while the solution is still running Load balancing lets large multi-server solutions maintain equilibrium of user load across different servers for optimal system performance

Additionally, you can also proactively make use of the following enhancements to improve the scalability of your solution:

  • Up to 64GB of addressable memory (EMA);
  • Application load balancing; and
  • Clustering supports for a virtually unlimited number of servers.

In part three, Techie Corner looks at the total cost of ownership for Linux.

Would you recommend this article?

Share

Thanks for taking the time to let us know what you think of this article!
We'd love to hear your opinion about this or any other story you read in our publication.


Jim Love, Chief Content Officer, IT World Canada

Featured Download

Featured Story

How the CTO can Maintain Cloud Momentum Across the Enterprise

Embracing cloud is easy for some individuals. But embedding widespread cloud adoption at the enterprise level is...

Related Tech News

Get ITBusiness Delivered

Our experienced team of journalists brings you engaging content targeted to IT professionals and line-of-business executives delivered directly to your inbox.

Featured Tech Jobs