Deploy Up to 5X More VMs on Any Storage, Lower Costs by Up to 50%
Atlantis ILIO USX™ is a significant breakthrough in virtualization technology, delivering a solution that unlocks the underutilized capacity of over $50 billion of deployed enterprise storage, similar to what VMware did for server virtualization.
Atlantis ILIO USX software gives IT the flexibility to get more out of their existing storage, even their older arrays, and to create new software-defined storage hybrid arrays, hyper-converged systems, and all-flash arrays by aggregating and pooling existing server SSDs, SAS, flash, and RAM together with shared SAN/NAS arrays.
Atlantis ILIO USX pools and optimizes server RAM, SAS, and/or Flash to create a highly scalable, hyper-converged platform using existing servers. Atlantis ILIO USX provides the flexibility to pool commodity local storage with RAM and/or flash across multiple server farms. By doing so, customers can seamlessly scale out their architecture to create a hyper-converged infrastructure and manage their applications without having to rip out their existing infrastructure.
Atlantis ILIO USX optimizes how storage is consumed by the application or VM by inserting a transparent software layer between the application and storage. The Atlantis ILIO USX software resides on the hypervisor platform as a set of virtual machines that can abstract any storage hardware into pools of virtual storage that can be combined together to form an Application Defined Storage Volume (ADS Volume). There are two types of storage pools:
These Capacity and Memory pools are combined together automatically to create Application Defined Storage Volumes (ADS Volume) with policy-based controls that apply the ideal combination of storage capacity, performance and availability for the application. The ADS Volume provides enterprise-class storage functionality including high-availability, data protection, thin provisioning and cloning. Multiple ADS Volumes can be created from a single Capacity and Performance Pool, enabling true application-centric storage for the first time.
The Atlantis ILIO USX platform applies multiple storage optimization technologies to boost the performance and increase the available storage capacity provided to the application. At the same time, Atlantis ILIO USX dramatically reduces the impact of application IO traffic on storage resource utilization (disk, storage controller, network).
Use with existing Shared Storage
Use Local SAS or SATA
Use RAM as Primary Storage
Optimized All-Flash Arrays
In-Memory Architecture — Atlantis ILIO USX can run VMs completely in server RAM to deliver high-speed, low latency runtime storage without re-architecting applications
IO Processing — Atlantis ILIO USX processes IO operations in real-time at the compute layer to lower latency and reduce network traffic
Inline De-duplication — Atlantis ILIO USX performs inline de-duplication in real-time on-the-wire with microsecond latency, eliminating up to 90% of storage IO traffic
Real-Time Compression — Atlantis ILIO USX compresses the optimized blocks In-Memory with microsecond latency
IO Blender Fix — Atlantis ILIO USX coalesces small random blocks generated by the hypervisor into larger sequential blocks, greatly improving storage access and efficiency
High Availability — Atlantis ILIO USX provides integrated high-availability and data protection to prevent application downtime.
Thin Provisioning — All Atlantis ILIO USX storage volumes are automatically thin provisioned with up to 10:1 consolidation.
Fast Clone — Atlantis ILIO USX can clone full VMs in as little as 4 seconds with no network or storage traffic.
Adaptive PSP is PSP plug-in that VMware developed to adaptively switch load-balancing strategies based on system load.
Moving forward, VMware are working to further enhance the adaptive logic by introducing a path scoring attribute that ranks different paths based on I/O latency, bandwidth, and other factors.
The score is used to decide whether a specific path should be used for different system I/O load conditions. Further, the logic decides the percentage of I/O requests that should be dispatched to a certain path, and could also combine the path score with I/O priorities, by introducing priority queuing within PSA.
Diskless Virtual Desktop Infrastructure (VDI) is the concept of using local server memory in combination with storage optimization software to store virtual desktop images instead of shared SAN/NAS or local SAS/SSD storage.
By storing virtual desktop images on the local memory of the hypervisor where the desktops execute, response time are faster than even the most expensive local SSD drives (MLC or SLC), cost less when combined with Atlantis ILIO, and increase reliability.
With existing VDI architectures, virtual desktop images are stored on either shared SAN/NAS storage or local SSD disks, which are costly, have limited IOPS for write-intensive VDI workloads, can have a limited lifespan and consume more power than memory.
Atlantis ILIO transparently offloads write operations before they reach the storage fabric.
A big change in vSphere 5.1 is the vMotion capabilities, where there is no need for shared storage between ESX hosts for Virtual Machines to migrate to and from. In previous versions of vSphere there was a requirement of shared storage.
See this in depth article on vSphere 5.1 vMotion to get a better understanding. http://frankdenneman.nl/vmotion/vsphere-5-1-vmotion-deepdive/
WHATS NEW in VMware vSphere 5.1
vSphere 5.1 is VMware’s latest release of its industry-leading virtualization platform. This new release contains the following new features and enhancements:
1) vRAM has been killed. vSphere is again licensed per CPU just as it was before vSphere 5.0. The licensing whitepaper is here.
2) Enhanced vMotion allows you to combine a vMotion and Storage vMotion into a single operation, without shared storage.
3) vSphere replication has been decoupled from SRM, and released as an available feature of every vSphere license from Essentials Plus through Enterprise Plus
5) New backup solution named vSphere Data Protection replacing VDR.
• Larger virtual machines – Virtual machines can grow two times larger than in any previous release to support even the most advanced applications. Virtual machines can now have up to 64 virtual CPUs (vCPUs) and 1TB of virtual RAM (vRAM).
• New virtual machine format – New features in the virtual machine format (version 9) in vSphere 5.1 include support for larger virtual machines, CPU performance counters and virtual shared graphics acceleration designed for enhanced performance.
• vSphere 5.1 introduces Flexible Space Efficiency (Flex-SE), a disk format to achieve the right balance of space efficiency and I/O throughput. This balance can be managed throughout the life cycle of a VM, from storage allocation (controlling the allocation block size) to how the blocks are managed after they are allocated (deleted blocks can be reclaimed). This feature enables the user to determine the right level of storage efficiency for a deployment. For example, you can use Flex-SE to optimize storage efficiency for virtual desktop infrastructure (VDI).
This feature will reclaim deleted storage in thin provisioned disk. Before vSphere 5.1 it was not possible to use the storage capacity freed when files are deleted inside the guest operating system.
• VMware increased the support for in guest clustering. In previous versions the support was limited to maximum 2 nodes running as a VM. Now support for up to 5 node Microsoft cluster using the Node Majority Model.
• vSphere 5,1 now supports a Storage vMotion up to 4 parallel disk migrations per operation. Support for 16 Gbps Fiber Channel has been added.
Support for boot from Software Fiber Channel over Ethernet ( FCoE ).
• Storage I/O Control (SIOC)
vSphere 5.1 improves SIOC functionality by automatically computing the best latency threshold for a datastore in lieu of using a default or user selected value. This latency threshold is determined by modeling, when 90% of the throughput value is achieved.
• vSphere 5.1 enables more granular latency measurement for I/O load balancing called “VMobservedLatency”. This is achieved by measuring the I/O request-response time between a VM and the datastore. In vSphere 5.0, latency was measured as theI/O request-response time between the host and the datastore.
• Advanced I/O Device Management
vSphere 5.1 introduces new commands for troubleshooting I/O adapters and storage fabrics. This enables diagnosis and querying of FC, and Fibre Channel over Ethernet (FCoE) adapters, providing statistical information that allows the administrator
to identify issues along the entire storage chain from the HBA to the ESXi, fabric and storage port.
•vSphere Distributed Switch – Enhancements such as
Network Health Check, Configuration Backup and Restore, Roll Back and Recovery, and Link Aggregation Control Protocol (LACP) support and deliver more enterprise-class networking functionality and a more robust foundation for cloud computing.
• Single-root I/O virtualization (SR-IOV) support
– Support for SR-IOV optimizes performance for sophisticated applications.
• vSphere vMotion® – Leverage the advantages of vMotion (zero-downtime migration) without the need for shared storage configurations. This new vMotion capability applies to the entire network.
• vSphere Data Protection – Simple and cost effective backup and recovery for virtual machines. vSphere Data Protection is a newly architected solution based EMC Avamar technology that allows admins to back up virtual machine data to disk without the need of agents and with built-in deduplication. This feature replaces the vSphere Data Recovery product available with previous releases of vSphere.
• vSphere Replication (separated from SRM) – vSphere Replication enables efficient array-agnostic replication of virtual machine data over the LAN or WAN. vSphere Replication simplifies management enabling replication at the virtual machine level and enables RPOs as low as 15 minutes.
• Zero-downtime upgrade for VMware Tools – After you upgrade to the VMware Tools available with version 5.1, no reboots will be required for subsequent VMware Tools upgrades.
• VMware vShield Endpoint™ – Delivers a proven endpoint security solution to any workload with an approach that is simplified, efficient, and cloud-aware. vShield Endpoint enables 3rd party endpoint security solutions to eliminate the agent footprint from the virtual machines, offload intelligence to a security virtual appliance, and run scans with minimal impact.
• vSphere Storage DRS™ and Profile-Driven Storage
– New integration with VMware vCloud® Director™ enables further storage efficiencies and automation in a private cloud environment.
• vSphere Auto Deploy™ – Two new methods for deploying new vSphere hosts to an environment make the Auto Deploy process more highly available then ever before. Management (with vCenter Server)
• vSphere Web Client –The vSphere Web Client is now the core administrative interface for vSphere. This new flexible, robust interface simplifies vSphere control through shortcut navigation, custom tagging, enhanced scalability, and the ability to manage from anywhere with Internet Explorer or Firefox-enabled devices.
• vCenter Single Sign-On – Dramatically simplify vSphere admin- istration by allowing users to log in once to access all instances or layers of vCenter without the need for further authentication.
• vCenter Orchestrator – Orchestrator simplifies installation and configuration of the powerful workflow engine in vCenter Server. Newly designed workflows enhance ease of use, and can also be launched directly from the new vSphere Web Client.
Download all the presentations – VMware Forum 2012 Manchester.
With an ESX host connected to a Clariion, there may be a time where you are experiencing connectivity issues. One of the first things you can do before you contact EMC support is to run the EMC Host Grab utility.
This utility will connect to your ESX host, and also the array. It will gather all the details it will need and will produce a zip file that can be uploaded to EMC Powerlink.
The EMC Primus article is “emc102335”.
Download the host grab utilities from:-
For UNIX hosts such as (AIX, HP-UX, Linux, Solaris, Tru64) ftp://ftp.emc.com/pub/emcgrab/Unix/
For ESX hosts ftp://ftp.emc.com/pub/emcgrab/ESX/
For ESXi hosts ftp://ftp.emc.com/pub/emcgrab/ESXi/
For Windows hosts ftp://ftp.emc.com/pub/emcgrab/Windows/
For Open VMS hosts ftp://ftp.emc.com/pub/emcgrab/vms/
From the web http://powerlink.emc.com
emcgrab.exe -host <host ip> -user <username> -password <password> -vmsupport –clariionip <sp a ip,sp b ip>
Example: – emcgrab.exe -host 10.0.0.1 -user root -password mypassword -vmsupport -clariionip 10.0.0.2, 10.0.0.3