IBM PureFlex – новейшее решение компании IBM

Итак, свершилось! На сайте IBM появилась информация о новейшем решении компании IBM – PureFlex System. Уже доступны несколько Redbooks на эту тему.

Вот небольшая выдержка из официальной информации

IBM Flex System™, a new category of computing and the next generation of Smarter Computing, is anchored by the IBM Flex System Enterprise Chassis. This platform offers intelligent workload deployment and management for maximum business agility. This chassis delivers high-speed performance complete with integrated servers, storage, and networking for multi-chassis management in data center compute environments. Furthermore, its flexible design can meet the needs of varying workloads with independently scalable IT resource pools for higher utilization and lower cost per workload. While increased security and resiliency protect vital information and promote maximum uptime, the integrated, easy-to-use management system reduces setup time and complexity, providing a quicker path to return on investment.

Introduction

IBM Flex System™, a new category of computing and the next generation of Smarter Computing, is anchored by the IBM Flex System Enterprise Chassis. This platform offers intelligent workload deployment and management for maximum business agility. This chassis delivers high-speed performance complete with integrated servers, storage, and networking for multiple chassis management in data center compute environments. Furthermore, its flexible design can meet the needs of varying workloads with independently scalable IT resource pools for higher utilization and lower cost per workload. While increased security and resiliency protect vital information and promote maximum uptime, the integrated, easy-to-use management system reduces setup time and complexity, providing a quicker path to return on investment. 

The IBM Flex System Enterprise Chassis is shown in Figure 1. 

Enterprise Chassis
Figure 1. The IBM Flex System Enterprise Chassis

Did you know?

IBM Flex System is the platform for IBM® PureFlex System. IBM PureFlex System combines advanced IBM hardware and software along with patterns of expertise and integrates them into three optimized configurations, Express, Standard, and Enterprise, that are simple to acquire and deploy so you get fast time to value for your solution.

Key features

The IBM Flex System Enterprise Chassis is a simple, integrated infrastructure platform that supports a mix of compute, storage, and networking resources to meet the demands of your applications. The solution is easily scalable with the addition of another chassis with the required nodes. With the IBM Flex System Manager, multiple chassis can be monitored from a single screen. The 14 node, 10U chassis delivers high-speed performance complete with integrated servers, storage, and networking. This flexible chassis is designed for a simple deployment now and to scale to meet your needs in the future.

Flexibility and efficiency

The 14 bays in the chassis allow the installation of compute or management nodes, with networking modules in the rear. A single chassis or a group of chassis can be fully customized to the specific needs of the computing environment. With support for POWER7® and Intel processor-based nodes, you can choose the architecture you need. IT can meet the needs of the business using a single system across multiple architectures and operating environments.

The system monitors and manages power usage on all major chassis components so you have total control over power consumption. The chassis supports N+N or N+1 redundant power supplies, configurable in either a single or three-phase power domain and an entirely passive mid-plane to meet your reliability needs. The power supplies are 80 PLUS Platinum-certified indicating high energy efficiency. The chassis design also optimizes cooling with cooling zones within the chassis. The system manages the fan modules based on node configuration within the chassis. So, the system can increase the speed of certain fan modules to cool potential hot spots, and use lower speeds for other fan modules where appropriate.

Easily scalable with simple administration

Because the IBM Flex System Enterprise Chassis is an all-in-one solution, it is designed for growth from a single chassis to many. Adding compute, storage, or networking capability is as simple as adding additional nodes, modules, or chassis. The simple, highly integrated management system allows you to use the Chassis Management Modules integrated into each chassis to administer a single chassis, or IBM Flex System Manager controls up to four chassis from a single panel.

Designed for multiple generations of technology

The IBM Flex System Enterprise Chassis is designed to be the foundation of your IT infrastructure now and into the future. Compute performance requirements are always on the rise and networking demands continue to grow with rising bandwidth needs and a shrinking tolerance for latency. The chassis is designed to scale to meet the needs of your future workloads, offering the flexibility to support current and future innovations in compute, storage, and networking technology.

Locations of key components and connectors

Figure 2 shows the front of the Enterprise Chassis.

Front of the Enterprise Chassis
Figure 2. Front of the IBM Flex System Enterprise Chassis

Figure 3 shows the rear of the Enterprise Chassis.

Rear of the Enterprise Chassis
Figure 3. Rear of the IBM Flex System Enterprise Chassis

Standard specifications

The following table lists the standard specifications. 

Table 1. Standard specifications

Components Specification
Model System x® ordering sales channel: 8721-A1x
Power Systems™ sales channel: 7893-92X
Form factor 10U rack-mounted unit.
Maximum number of compute nodes supported 14 half-wide (single bay), seven full-wide (two bays), or three double-height full-wide (four bays). Mixing is supported.
Chassis per 42U rack Four.
Nodes per 42U rack 56 half-wide, or 28 full-wide.
Management One or two Chassis Management Modules (CMMs), for basic chassis management. Two CMMs form a redundant pair; One CMM standard in 8721-A1x. The CMM interfaces with the integrated management module (IMM) or Flexible Service Processor (FSP) integrated in each compute node in the chassis. Optional IBM Flex System Manager management appliance for comprehensive management, including virtualization, networking, and storage management.
I/O architecture Up to eight lanes of I/O to an I/O adapter card, with each lane capable of up to 16 Gbps bandwidth. Up to 16 lanes of I/O to a half-wide node with two adapters. A wide variety of networking solutions, including Ethernet, Fibre Channel, Fibre Channel over Ethernet (FCoE), and InfiniBand.
Power supplies Six 2500-watt power modules that provide N+N or N+1 redundant power; Two standard in model 8721-A1x. Power supplies are 80 PLUS Platinum-certified and provide 95% efficiency at 50% load and 92% efficiency at 100% load. Power capacity of 2500-watt output rated at 200VAC. Each power supply contains two independently powered 40 mm (1.57 in) cooling fan modules.
Fan modules Ten fan modules (eight 80 mm (3.14 in) fan modules and two 40 mm (1.57 in) fan modules); Four 80 mm and two 40 mm fan modules standard in model 8721-A1x.
Dimensions Height: 440 mm (17.3 in)
Width: 447 mm (17.6 in)
Depth:

  • 800 mm (31.5 in) (measured from front bezel to rear of chassis)
  • 840 mm (33.1 in) (measured from node latch handle to the power supply handle)
Weight Minimum configuration: 96.62 kg (213 lb)
Maximum configuration: 220.45 kg (486 lb)
Declared sound level 7.5 bels
Temperature Operating air temperature 5°C to 40°C
Electrical power Input power: 200 – 240 V ac (nominal), 50 or 60 Hz
Minimum configuration: 0.51 kVA (two power supplies)
Maximum configuration: 13 kVA (six power supplies)
Power consumption 12,900 watts maximum


The IBM Flex System Enterprise Chassis model 8721-A1x ships with the following items:

  • One Chassis Management Module
  • Two 2500 W power supplies
  • Four 80 mm Fan Modules
  • Two 40 mm Fan Modules
  • Four Power Supply Fillers
  • One Console Breakout Cable
  • Two C19 to C20 two-meter power cables
  • One Rack Mount Kit

Supported compute nodes

The following table lists the compute nodes that are supported in the IBM Flex System Enterprise Chassis. The table also lists the maximum number installable. 

Table 2. Supported compute nodes and maximum quantities


Description

Machine type

Maximum number of 
nodes per chassis

IBM Flex System x240 Compute Node

8737

14

IBM Flex System p260 Compute Node

7895-22X

14

IBM Flex System p460 Compute Node

7895-42X

7

IBM Flex System Manager Node

8731-A1x

1 (2 per four chassis)


See IBM® ServerProven® at the following web address for the latest information about the servers supported in the IBM Flex System Enterprise Chassis:
http://ibm.com/servers/eserver/serverproven/compat/us/

Supported I/O modules

The IBM Flex System Enterprise Chassis has four high-speed switch bays that are capable of supporting a variety of I/O architectures. 

The switches are installed in switch bays in the rear of the IBM Flex System Enterprise Chassis as shown in the following figure. Switches are normally installed in pairs (bays 1 & 2, and bays 3 & 4), because I/O adapter cards installed in the compute nodes route to two switch bays for performance and redundancy.

Location of the switch bays in the Enterprise Chassis
Figure 4. Location of the switch bays in the IBM Flex System Enterprise Chassis

The following figure shows one of the available switches, the IBM Flex System Fabric EN4093 10Gb Scalable Switch. It offers the following interface ports: 

  • Up to 42 internal 10Gb ports, three to each half-wide compute node bay (not shown in the figure)
  • Up to 14 external 10Gb ports with SFP+ connectors (license upgrades enable all ports)
  • Up to 2 external 40Gb uplink ports with QSFP+ connectors (license upgrades enable these ports)


Fabric EN4093 10Gb Scalable Switch
Figure 5. IBM Flex System Fabric EN4093 10Gb Scalable Switch 

The following table lists the switches that are available for the chassis. 

Table 3. I/O modules and upgrades

Description Part number System x
feature code
Power Systems
feature code
Ethernet switches
IBM Flex System Fabric EN4093 10Gb Scalable Switch 49Y4270 A0TB 3593
IBM Flex System EN4091 10Gb Ethernet Pass-thru 88Y6043 A1QV 3700
IBM Flex System EN2092 1Gb Ethernet Scalable Switch 49Y4294 A0TF 3598
Fibre Channel switches
IBM Flex System FC5022 16Gb SAN Scalable Switch 88Y6374 A1EH 3770
IBM Flex System FC5022 24-port 16Gb ESB SAN Scalable Switch 90Y9356 A1EJ 3771
IBM Flex System FC3171 8Gb SAN Switch 69Y1930 A0TD 3595
IBM Flex System FC3171 8Gb SAN Pass-thru 69Y1934 A0TJ 3591
InfiniBand switches
IBM Flex System IB6131 InfiniBand Switch (QDR/FDR) 90Y3450 A1EK

3699

I/O architecture

Each half-wide compute node (such as the IBM Flex System x240 Compute Node) has two adapter slots, and each full-wide compute node (such as the IBM Flex System x460 Compute Node) has four adapter slots. The adapter slots in each compute node route through the chassis midplane to the switch bays. The architecture supports up to eight ports per adapter although currently only two-port and four-port adapters are available. 

The following figure shows how two-port adapters are connected to switches installed in the chassis. 

Logical layout of the interconnects between two-port I/O adapters and I/O modules
Figure 6. Logical layout of the interconnects between two-port I/O adapters and I/O modules

A four-port adapter doubles the connections between each adapter and switch pair (for example, a four-port adapter in A1 in each compute node routes two connections to switch 1 and two connections to switch 2). 

The following figure shows how four-port adapters are connected to switches that are installed in the chassis.

Logical layout of the inter-connections between four-port I/O adapters and I/O modules
Figure 7. Logical layout of the inter-connections between four-port I/O adapters and I/O modules

To make use of all four ports of a four-port card, you must install two switches in the chassis, and each switch must have 28 internal ports enabled. The IBM Flex System Fabric EN4093 10Gb Scalable Switch with license upgrade 1 enables 28 internal ports, for example. Similarly, when 6-port cards become available, two switches must be installed, and 42 internal ports must be enabled. The IBM Flex System Fabric EN4093 10Gb Scalable Switch with license upgrade 2 enables a total of 42 internal ports, for example. 

The following table shows the connections between adapter slots in the compute nodes to the switch bays in the chassis.

Table 4. Adapter to I/O bay correspondence

I/O adapter slot
in the server
Port on the adapter Corresponding I/O module bay
in the chassis
Slot 1 Port 1 Module bay 1
Port 2 Module bay 2
Port 3 (for 4-port cards)* Module bay 1
Port 4 (for 4-port cards)* Module bay 2
Slot 2 Port 1 Module bay 3
Port 2 Module bay 4
Port 3 (for 4-port cards)* Module bay 3
Port 4 (for 4-port cards)* Module bay 4
Slot 3
(full-wide compute nodes only)
Port 1 Module bay 1
Port 2 Module bay 2
Port 3 (for 4-port cards)* Module bay 1
Port 4 (for 4-port cards)* Module bay 2
Slot 4
(full-wide compute nodes only)
Port 1 Module bay 3
Port 2 Module bay 4
Port 3 (for 4-port cards)* Module bay 3
Port 4 (for 4-port cards)* Module bay 4

* To make use of all four ports of a four-port adapter, the switch must have 28 internal ports enabled, and two switches must be installed in the bays as indicated.

Chassis Management Module

The Chassis Management Module (CMM) provides single-chassis management. The CMM is used to communicate with the management controller in each compute node (IMMv2 in Intel processor-based compute nodes and FSP in POWER7® processor-based compute nodes) to provide system monitoring, event recording and alerts, and to manage the chassis, its devices, and the compute nodes. 

The chassis supports up to two CMMs. If one CMM fails, the second CMM can detect its inactivity and activate itself to take control of the system without any disruption. The CMM is central to the management of the chassis and is required in the Enterprise Chassis.

The following figure shows the Chassis Management Module. See Figure 3 for the location of the CMM in the chassis.

Chassis Management Module
Figure 8. Chassis Management Module 

The CMM provides these functions:

  • Power control
  • Fan management
  • Chassis and compute node initialization
  • Switch management
  • Diagnostics: chassis, I/O options, and compute nodes
  • Resource discovery and inventory management
  • Resource alerts and monitoring management
  • Chassis and compute node power management
  • Security policy management
  • Role-based access control
  • Support for up to 84 local CMM user accounts
  • Support for up to 32 simultaneous sessions


The CMM has the following connectors:

  • USB connection. This connection can be used for insertion of a USB media key for tasks, such as firmware updates.
  • 10/100/1000 Mbps RJ45 Ethernet connection to connect to a management network. The CMM can be managed via this Ethernet port.
  • Serial port (mini-USB) for local command-line interface (CLI) management. Use serial cable 90Y9338 for connectivity.


The CMM has the following light-emitting diodes (LEDs) that provide the following information:

  • Power-on LED
  • Activity LED
  • Error LED
  • Ethernet port link and port activity LEDs


The CMM also incorporates a reset button, which, when pressed, resets the CMM back to its default condition. It has two functions, depending on how long the button is pressed:

  • When pressed for less than 5 seconds, the CMM restarts.
  • When pressed for more than 5 seconds (for example, 10 or 15 seconds), the CMM configuration is reset to the manufacturing defaults, and then the CMM restarts.


The CMM supports a web-based graphical user interface (GUI) that provides a way to perform CMM functions within a supported web browser. You can also perform management functions through the CMM command-line interface (CLI). Both the web-based GUI and the CLI are accessible via the single RJ45 Ethernet connector on the CMM or from any other system that is connected to the same (management) network.

The CMM has the following default static IPv4 address. By default, the CMM is configured to respond to Dynamic Host Configuration Protocol (DHCP) first before using its static IPv4 address: 

  • IP address: 192.168.70.100
  • Subnet: 255.255.255.0
  • User ID: USERID (all capital letters)
  • Password: PASSW0RD (all capital letters, with a zero instead of the letter O)


The CMM does not have a fixed static IPv6 IP address, by default. Initial access to the CMM in an IPv6 environment can be performed by either using the IPv4 IP address or the IPv6 link-local address. The IPv6 link-local address is automatically generated based on the Media Access Control (MAC) address of the CMM. 

The CMM is the key component enabling the integrated management network. Internally, the CMM has a multiple port L2 1Gigabit Ethernet switch with dedicated links to all 14 node bays, all four switch bays, and the second CMM, if installed. These connections are all point-to-point, ensuring dedicated bandwidth. The 1GbE links are full-duplex, fixed speed (not auto-negotiate) links. The 1 GbE management network is only accessible by each node’s management controller (IMMv2 or FSP), each switch module’s management interfaces, the CMM, and the IBM Flex System Manager (FSM) management appliance. This design permits the separation of the management network from the data network. 

The CMM has a high-security policy that is enabled by default, which means that the following policies are enabled by default:

  • Strong password policies with automatic validation and verification checks
  • Required update of the default passwords after the initial setup
  • Only secure communication protocols, such as SSH and SSL. Unencrypted protocols, such as HTTP, Telnet, and SNMPv1, are disabled.
  • Certificates to establish secure, trusted connections for applications that run on the management processors

IBM Flex System Manager

IBM Flex System Manager (FSM) is a systems management solution that offers flexible integrated systems management across compute, storage, and networking resources in the IBM Flex System infrastructure. It is designed to help you get the most out of your IT installation while automating repetitive tasks. It can significantly reduce the number of manual navigational steps for typical management tasks. From simplified system setup procedures with wizards and built-in expertise to consolidated monitoring for all of your physical and virtual resources, FSM provides core management functionality along with automation so that you can focus your efforts on business innovation.

The IBM Flex System Manager compute node, 8731-A1x, provides configuration and management support locally or remotely, for up to four chassis, their devices, and the compute nodes. You can install up to two FSMs in each four-chassis configuration (one FSM in two of the four chassis) for redundancy, which enables the system to continue to operate without disruption if one fails. 

The following figure shows the IBM Flex System Manager.

IBM Flex System Manager
Figure 9. IBM Flex System Manager

The IBM Flex System Manager drives efficiency and cost savings in the data center. It provides a pre-integrated and virtualized management environment across servers, storage, and networking that is easily managed from a single pane of glass. It offers the following key features:

  • Single focus point for seamless multichassis management provides an instant resource-oriented view of the chassis and chassis resources for compute nodes.
  • Design reduces the number of interfaces, steps, and clicks that it takes to manage IT resources.
  • Capability to intelligently manage and deploy workloads based on resource availability and predefined policies.
  • Management of events and alerts to increase system availability and reduce downtime.
  • It helps to drive energy savings and reduce operational costs.


The IBM Flex System Manager offers these main features:

  • Monitoring and problem determination:
    • A real-time multichassis view of hardware components with overlays for additional information
    • Automatic detection of issues in your environment through an event setup that triggers alerts and actions
    • Identification of changes that might affect availability
    • Server resource utilization by virtual machine or across a rack of systems
  • Hardware management:
    • Automated discovery of physical and virtual servers and interconnections, applications, and supported third-party networking
    • Inventory of hardware components
    • Chassis and hardware component views, including hardware properties, component names and part numbers, firmware levels, energy usage and power consumption, utilization rates, and temperature readings
  • Network management:
    • Management of network switches
    • Discovery, inventory, and status monitoring of switches
    • Graphical network topology views
    • Support for KVM, pHyp, and VMware virtual switches, and physical switches
    • Virtual LAN (VLAN) configuration of switches
    • Integration with server management
    • Per-virtual machine network usage and performance statistics provided to VMControl
    • Logical views of servers and network devices grouped by subnet and VLAN
  • Storage management:
    Storage can also be consolidated onto the FSM platform, or users can virtualize the integrated enterprise storage with other multivendor storage systems within the existing environment, and then control all virtualized storage resources from the IBM Flex System platform.
  • Virtualization and workload management:
    • Consolidate physical infrastructure or applications onto fewer servers
    • Deploy new or existing workload onto Linux, Microsoft Windows, or competitive platforms
    • Upgrade current rack or blade infrastructure to an integrated environment
    • Optimize the performance of applications by migrating them to Flex System Manager
    • Move into advanced stages of virtualization or a private cloud deployment
    • Deploy new workloads rapidly
    • Optimize or upgrade high-value, noninfrastructure workloads, such as enterprise resource planning (ERP), customer relationship management (CRM), data warehousing, or analytics
  • Additional features:
    • Resource-oriented chassis map provides instant graphical view of chassis resources, including nodes and I/O modules
    • Remote console, including remote keyboard, video, mouse (KVM) session, remotely mounted optical and USB media, and remote power control
    • Hardware detection and inventory creation
    • Firmware compliance and updates
    • Automatic detection of hardware failures and alerting. Capability to take corrective action, including automatically notifying IBM of problems to escalate problem determination
    • Health status, such as CPU utilization, on all hardware devices from a single chassis view
    • Administrative capabilities, such as setting up users within profile groups, assigning security levels, and security governance

Подробности тут 

http://www.redbooks.ibm.com/abstracts/tips0863.html?Open

Оставьте комментарий