There are several enhancements in the new vSphere 5.5 around scalability, performance, availability, and user interface.
- vSphere configuration maximum increases
- Virtual Machine Compatibility ESXi 5.5 (vHW10)
- vCenter Appliance maximums
- Expanded vGPU and GP-GPU support
- Graphic acceleration for Linux guests
- Enhancements for CPU C-States
- Hot-pluggable SSD PCIe devices
- Support for Reliable Memory
- Support for database clustering technologies
- Increased platform support
- Enhanced usability Experience
The vSphere configuration maximum increases make it so there isn't a workload that the vSphere platform cannot virtualize. From Tier1 enterprise applications to next-generation applications, they can all take advantage of the new configuration maximums without any issues. In addition, the new Virtual Machine Compatibility 10 will support a new virtual-SATA controller that can support up to 120 disk devices which is a 2x increase from vSphere 5.1. At VMworld, VMware was giving away bumper stickers that read vSphere Loves Applications. Indeed!
vCenter Server Appliance has been gaining popularity and now with the support for a higher scale utilizing the embedded database (vPostgres), the vCenter server appliance can now manage up to 500 vSphere hosts / 5000 virtual machines which simplifies the infrastructure required to deploy and operate vCenter server.
vSphere 5.1 was the first release of vSphere to provide support for hardware-accelerated 3D graphics (vGPU) inside of a virtual machine. This support was limited to only NVIDIA based GPUs with this first release. With vSphere 5.5, vGPU support has been expanded to include both Intel and AMD based GPUs. Virtual machines with graphic-intensive workloads or applications that that require hardware-based vGPUs can now take now advantage of additional vGPU vendor make and models. Intensive workloads/apps that require 3D graphics can now be virtualized without impact to performance and multiple GPU vendors are supported to provide customers with choice.
In earlier versions of vSphere, the Balanced Policy for Host Power Management leveraged only P-States (Performance State), which kept the processor running at a lower frequency and voltage. In vSphere 5.5, deep C-States (Processor Power State) will also used as well to provide additional power savings. Another potential benefit to lower power consumption is the potential increased performance as Turbo Mode frequencies on Intel chipsets can be reached quicker as other CPU cores in the physical package are in deep C-States.
Here are the the different CPU power states according to Wikipedia:
The ACPI specification defines the following four Global "Gx" states and six Sleep "Sx" states for an ACPI-compliant computer-system:
- G0 (S0): Working. "Awaymode" is a subset of S0, where monitor is off but background tasks are running.
- G1, Sleeping. Divided into four states, S1 through S4:
- S1: All the processor caches are flushed, and the CPU(s) stops executing instructions. The power to the CPU(s) and RAM is maintained. Devices that do not indicate they must remain on, may be powered off.
- S2: CPU powered off. Dirty cache is flushed to RAM.
- S3: Commonly referred to as Standby, Sleep, or Suspend to RAM (STR). RAM remains powered.
- S4: Hibernation or Suspend to Disk. All content of the main memory is saved to non-volatile memory such as a hard drive, and is powered down.
- G2 (S5), Soft Off: G2/S5 is almost the same as G3 Mechanical Off, except that the PSU still supplies power, at a minimum, to the power button to allow return to S0. A full reboot is required. No previous content is retained. Other components may remain powered so the computer can "wake" on input from the keyboard, clock, modem, LAN, or USB device.
- G3, Mechanical Off: The computer's power has been totally removed via a mechanical switch (as on the rear of a PSU). The power cord can be removed and the system is safe for disassembly (typically, only the real-time clock continues to run - using its own small battery).
Furthermore, the specification defines a Legacy state: the state on an operating system which does not support ACPI. In this state, the hardware and power are not managed via ACPI, effectively disabling ACPI.
The device states D0–D3 are device-dependent:
- D0 Fully On is the operating state.
- D1 and D2 are intermediate power-states whose definition varies by device.
- D3 Off has the device powered off and unresponsive to its bus.
- D3 Hot & Cold: The D3 state is further divided into D3 Hot (has aux power), and D3 Cold (no power provided). A device in D3 Hot state can assert power management requests to transition to higher power states.
The CPU power states C0–C3 are defined as follows:
- C0 is the operating state.
- C1 (often known as Halt) is a state where the processor is not executing instructions, but can return to an executing state essentially instantaneously. All ACPI-conformant processors must support this power state. Some processors, such as the Pentium 4, also support an Enhanced C1 state (C1E or Enhanced Halt State) for lower power consumption.
- C2 (often known as Stop-Clock) is a state where the processor maintains all software-visible state, but may take longer to wake up. This processor state is optional.
- C3 (often known as Sleep) is a state where the processor does not need to keep its cache coherent, but maintains other state. Some processors have variations on the C3 state (Deep Sleep, Deeper Sleep, etc.) that differ in how long it takes to wake the processor. This processor state is optional.
Additional states are defined by manufacturers for some processors. For example, Intel's Haswell platform has states up to C10, where it distinguishes core states and package states.
While a device or processor operates (D0 and C0, respectively), it can be in one of several power-performance states. These states are implementation-dependent. Though, P0 is always the highest-performance state; with P1 to Pn being successively lower-performance states, up to an implementation-specific limit of n no greater than 16.
P-states have become known as SpeedStep in Intel processors, as PowerNow! or Cool'n'Quiet in AMD processors, and as PowerSaver in VIA processors.
- P0 max power and frequency
- P1 less than P0, voltage/frequency scaled
- P2 less than P1, voltage/frequency scaled
- Pn less than P(n-1), voltage/frequency scaled
The ability to hot-swap traditional storage devices such as a SATA/SAS hard disk on a running vSphere host has been a huge benefit to system administrators in reducing the amount of downtime for virtual machine workloads. As solid-state disks (SSD) are becoming more prevalent in the Enterprise datacenter, this same capability has been expanded to support SSD devices. Similar to SATA/SAS hard disk, users will now be able to hot add or remove an SSD device while a vSphere host is running and the underlying storage stack will be aware of the operation.
The most critical component to the vSphere ESXi Hypervisor is the VMkernel, which is a purpose built Operating System to run Virtual Machines. Since the vSphere ESXi Hypervisor runs directly in memory, an error in the hypervisor can potentially crash it and the virtual machines running on the host. To provide greater resiliency and guard against memory errors, the vSphere ESXi Hypervisor can now take advantage of a CPU hardware feature known as Reliable Memory. This works where a region of memory is reported up to the vSphere ESXi Hypervisor by the hardware as being more “reliable”. This information is then used to optimize the placement of the VMkernel and other critical components such as the init thread, hostd and the Watchdog process and helps guard against memory errors.
With vCenter 5.5 they officially reintroducing support for cluster technologies on the backend databases (VC,SSO etc) with Oracle and MSSQL solutions fully tested and now fully supported.
vSphere Web Client now supports the OSX platform with the client integration tools plugin adding greater platform support for accessing your vSphere environment. Using the vSphere Web client has also become more useable with the introduction of drag and drop to simplify bulk operations, display filters to quickly narrow displayed objects to those that matter most and recent items visible to return to previously selected objects without having to use the typical navigation.