Blog

Press Review #2

Oct 04, 2011 | 8 minutes read
Share this:

Tags: Press

Here is a little press review around Oracle technologies, and Solaris in particular:

The Fast Reboot feature enables you to reboot an x86 based system, bypassing the firmware and boot loader processes. Fast Reboot implements an in-kernel boot loader that loads the kernel into memory and then switches to that kernel, so that the reboot process occurs within seconds. This feature is implemented on both 32-bit and 64-bit kernels.

If you are running Oracle Solaris 11 Express, Fast Reboot is enabled by default on the x86 platform, without the need to use the -f option with the reboot command. The Fast Reboot feature of Oracle Solaris, previously introduced on the x86 platform, is now supported on the SPARC platform. The integration of Fast Reboot on the SPARC platform enables the -f option to be used with the reboot command to accelerate the boot process by skipping certain POST tests.

On both the x86 and SPARC platforms, Fast Reboot is managed through SMF and implemented through a boot configuration service, svc:/system/boot-config. The boot-config service provides a means for setting or changing the default boot configuration parameters. When the config/fastreboot_default property is set to true, the system performs a fast reboot automatically, without the need to use the reboot -f command. By default, this property value is set to true on the x86 platform and to false on the SPARC platform.

With its acquisition of Ksplice, Oracle is now in the position of being the only enterprise Linux provider that can offer zero downtime updates. In its customer letter Oracle announces that it plans to make the Ksplice technology a standard feature of Oracle Linux Premier Support. The letter also mentions that Oracle does not plan to support the use of Ksplice technology with Red Hat Enterprise Linux or SUSE Enterprise Linux. Rather, the Oracle Linux Premier Support subscription applies to Unbreakable Enterprise Kernel.

The Edison Group's white paper, "Oracle x86 Infrastructure: The Optimized Stack: Reducing Total Cost of Ownership through Vertical Integration" examines the cost structures across a range of system sizes and deployments for the core x86 system stack by comparing Oracle's integrated complete infrastructure with alternatives from HP and HP, all deployed with Red Hat Enterprise Linux and VMware vSphere, both together and separately. Among the findings is that TCO with an Oracle solution is as much as 57 percent lower than comparable deployments. Edison evaluated two, four and eight socket systems over three and five year periods using the Oracle Sun Fire X4170 M2 server for the two socket study and the Sun Fire X4470 M2 for the four socket system. For the eight socket study, Edison Group used the Sun Fire X4800 M2, and for a ten two socket study of blades with networking Edison used the Sun Blade X6270 M2 server module in a Sun Blade 6000 chassis with Sun Blade 6000 10 gigabyte (GB) switched NEM 24p. Equipment from IBM and HP was comparable in every case.

The white paper concludes that, "By engineering the entire infrastructure with service and support in mind, Oracle can deliver lower TCO in the design and operation of its system, in the ease of deployment enabled by VM Templates and Validated Configurations ... and in the efficiency and effectiveness of its ... Premier Support package."

Feature highlights in the Oracle Solaris Studio 12.3 Beta Release include:
Application Performance\

  • Compiler and library optimizations for the newest SPARC and x86 Oracle Sun Enterprise servers
    Application Observability\
  • New application quality tool, Code Analyzer, to identify application vulnerabilities,
    including memory, code coverage and other errors\
  • Improved performance analysis tooling for multi-core applications
    Developer Productivity\
  • Compile time performance enhancements\
  • Updated support for key open source runtime libraries, including BOOST and Apache C++\
  • Feature enhancements throughout toolchain to simplify multi-core development\
  • New support for Oracle Database development with Pro*C support within IDE
    Platform support for Oracle Solaris and Oracle Linux\
  • Includes Oracle Solaris 11 Express and Oracle Linux\
  • New IDE support on Windows enables remote development

This is the third article in a series highlighting best practices for software updates in Oracle Solaris 11 Express. The first article introduced the IPS software packaging model and highlighted best practices for creating a new Boot Environment (BE) before performing an update. The second article discussed the Time Slider and auto-snapshot services, describing how to initialize and use these services to periodically snapshot BEs and other ZFS volumes.

This third article dives more deeply into the topic of software updates, exploring the process of updating an Oracle Solaris 11 Express system configured with zones. This topic is especially pertinent since zones in this release differ somewhat from those in Oracle Solaris 10, as does the software upgrade process for zoned systems.

...that is the question.

Ever since the introduction of deduplication into ZFS, users have been divided into two camps: One side enthusiastically adopted deduplication as a way to save storage space, while the other remained skeptical, pointing out that dedupe has a cost, and that it may not be always the best option.

Let's look a little deeper into the benefits of ZFS deduplication as well as the cost, because ultimately it boils down to running a cost/benefit analysis of ZFS deduplication. It's that simple.

Oracle Solaris 11 Express is distributed in several formats: a hands-free server based format that's used for automatic installation, an interactive installer format that only has console access, and an interactive graphical installer that includes a full desktop environment.

This article focuses on the third format with the full desktop environment, although most of the concepts discussed in this article will apply to any of the Oracle Solaris 11 Express formats. The graphical format is officially known as the LiveCD. This means that Oracle Solaris can be booted into RAM, causing zero impact on your existing operating system. Once it is loaded, you are then free to experiment with Oracle Solaris to determine whether it is something that you would like to install.

The LiveCD is not intended for long-term use. For example, any changes that you make to the system will be lost when the system is shut down. Therefore, the next logical step is to install Oracle Solaris on the system, which the LiveCD makes easy by placing an Install Oracle Solaris icon right on the desktop. But before we head down that road, let's step back a bit and consider the installation options.

For years I, and many others, have been tuning TCP, UDP, IP, and other aspects of the Solaris network stack with ndd(1M). The ndd command is documented, however, most of the tunables were really private interface implementations, subject to change, and lacked documentation in many cases. Also, ndd does not show the default values, nor the possible values or ranges.

That is changing with Solaris 11 Express. A new command ipadm(1M) allows persistent and temporary (with the -t option) setting of key tunable values. This is a major improvement over ndd, where it is customary to create an /etc/rc2.d/S69ndd or similar script to set the parameter on every reboot. Another benefit is that ipadm shows the default value and the values that the property can be set to.

The ipadm has many features to configure the IP settings of interfaces. This blog entry focuses on how ipadm replaces ndd. Note that ipadm only supports the IP, TCP, UDP, SCTP, and ICMP protocols. Other protocols such as ipsecah and keysock still required the use of ndd.

Oracle made a maintenance release of Oracle VM VirtualBox version 4.1.2.

This release fixes lots of stuff including a significant problem on AMD Fusion CPUs, and also adds Linked Clone support to the VirtualBox Manager (GUI) which we couldn't get done in time for 4.1.0.

Performance engineers often look at improving application performance by getting the compiler to produce more efficient binaries from the same source. This is done by changing what compiler options are used. In this modern era of Open Source Software, you can often get your hands on a number of binary distributions of an application, but if you really want to roll your sleeves up, the source is there, just waiting to be compiled with the latest compiler and optimizations.

Now, it might be useful to have as a reference the compiler version and flags that were originally used on the binary distribution you tried out, or you just might be interested to know. Read on for details on the forensic tools.