HPE SGI Management Suite: технические характеристики и документация

HPE SGI Management Suite v3.5 

HPE SGI Management Suite efficiently deploys and manages the HPE SGI 8600 system with one comprehensive toolset. The software offers an array of features, including the ability to initiate management actions, monitor essential system health metrics, and improve overall power efficiency. By combining the discovery of nodes with multicast technology, users can significantly shorten the time required to complete provisioning.

HPE SGI Management Suite address system management tasks starting with provisioning, to the daily tasks of system health monitoring, proactive problem resolution, updating software images and resource management. The software offers the following key features:

• Multicast provisioning
• Version-control image management
• The ability to monitor system logs and broadcast alerts
• Hardware event tracking
• Power management
• 24x7 monitoring and change detection
• Memory error logging
• Predictive failure analysis for memory
• Page retirement for memory
• Supported on HPE SGI 8600 systems with SUSE® Linux Enterprise Server 12, Red Hat Enterprise Linux 7, and CentOS 7

What's New 

Provisioning

• Faster identification of network issues related to multicast provisioning
• Improved reliability of multicast transfers by reducing wait time for failing nodes  

API

• New API tools, including allowing workload managers to access the power on/ off status for a single node or a set of nodes  

Extended Support

• Gen10 system support for the HPE SGI 8600 (which now supports HPE FlexNetwork 5510 switches)
• Support for additional operating system versions:  SLES 12 SP3, RHEL 7.4 and CentOS 7.4

Models and Licensing

HPE SGI Management Suite 1 Socket 3yr Support LTU

Q2A47A

NOTE:  One license per socket.

NOTE:  Includes three (3) years of HPE Pointnext support.

NOTE:  Please contact HPE Pointnext for 4- and 5-year support options.

HPE SGI Management Suite Media Kit

Q2A36A

NOTE:  One media kit per solution.

HPE SGI Management Suite FIO

Q2A41A

NOTE:  For factory installation only; this PN does not include the software license. Please order with PN Q2A47A.

NOTE:  Order one per node.

Distribution Media and Software Documentation

HPE SGI Management Suite software and documentation is available on a single DVD.  Please select HPE SGI Management Suite Media Kit (PN Q2A36A) to order the DVD. 

HPE SGI Management Suite is also available for download. Upon receiving their order, customers will receive a physical document, which includes the website URL to access the software, as well as additional license information. Software will be available for download via the Software Updates and Licensing Portal. 

Patches may be required and will be delivered via the Software Download Repository. Release notes will be made available for any patches issued to customers. 

Factory installation is available for the HPE SGI 8600 and Apollo 40 systems. Additional platforms are not supported at this time. Please select PN Q2A41A for factory installation. 

Customers may download user guides and other documentation at hpe.com/software/sms.

Provisioning

HPE SGI Management Suite combines the discovery of cluster nodes and multicast provisioning to reduce the time time required to setup a large scale cluster. HPE SGI 8600 systems are factory installed with a Linux operating system and middleware software. The system configuration file used for hardware discovery is included with the software installed on the system.

The provisioning process uses the system configuration file to locate each server in the clustered system that requires a copy of the operating system. Multicast technology provisions servers in parallel—enabling an operating system image to be installed on several servers in the same amount of time as a single server. Multicast provisioning significantly decreases the amount of downtime during system maintenance periods. Systems can be installed or updated in a matter of minutes instead of days.

Version-Controlled Image Management

Node specific software resides on the admin node within the HPE SGI 8600 system. When a system administrator installs and configures the cluster software, the installer pushes the node-specific software to each node in the cluster. Over time, these software images may require modification. If the images are modified frequently, the image repository will contain several different versions of an image and become difficult to manage.

To solve this problem, HPE SGI Management Suite includes a version control system, which archives, tracks, and manages the various versions of an image.

Images are stored in a RPM format, allowing them to be easily deployed onto different servers within the system. If problems arise after an operating system upgrade, the system can be returned to a known working state. Updates to images and payloads are stored as changes to the originals to significantly save on disk space, while maintaining a full revision history. Changes to the image management archive are tracked by date and by who made the change, enabling accountability for revision activities.

System Monitoring

HPE SGI Management Suite includes a hardware event tracker that receives SNMP events from the IPMI management controller on the hardware system. When a non-critical event occurs, the hardware event tracker records the event in a system log. If a system condition becomes too low or too high for its threshold, an email alert is generated detailing the crticial event. The email is then sent to the system administrator. The following are examples of critical system events that cause an alert:

• Ambient air temperature is too high
• Low voltage
• Power supply failure
• Loss of redundant power supply
• Fan speed is unable to attain a critical threshold or a loss of fan redundancy
• Board processor modules that exceed a critical temperature threshold
• Memory uncorrectable errors

System administrators may also view system metrics via Ganglia, a scalable, distributed monitoring system that displays real-time histograms of system metrics. Each HPE SGI 8600 compute node acts as a single monitoring source that sends its statistics to the leader node. After collecting the data, the leader node forwards aggregated rack statistics (along with its own data) to the admin node. The admin node collects data from all of the leader nodes and presents the cluster-wide metrics. This model enables HPE to scale-out Ganglia to very large cluster deployments.

By default, Ganglia gathers the following metrics. Please review the table below to understand if these metrics are captured at the admin, leader or compute node level.

Remote Services

HPE SGI Management Suite also includes an option for Remote Services, which also monitors the system logs for alerts sent directly to HPE Pointnext. In this scenario, HPE Pointnext is responsible for analysis, action and communication with the customer system administrator. This service enhances the HPE support experience by providing 24x7 remote monitoring and data gathering on the customer's system. Remote Services provides a connection to Technical Support from HPE PointNext, which includes secure file transfer and optional remote access to the system.

HPE SGI Management Suite leverages Nagios, a web-based monitoring tool for networks and clusters, to monitor the health of the specified cluster nodes and services, notify a specified audience by email or SMS if a critical event occurs, and gather and display statistics.  Nagios plays a key role in the 24x7 monitoring offered to HPE SGI Management Suite users.

The open source Nagios architecture is also extensible so that it is easy to add new plug-ins to support additional reporting mechanisms.

Memory Error Handling

HPE SGI Management Suite includes memory error handling software, which collects memory error information from all memory residing in the hardware system, node or cluster. System performance can be affected by memory errors and this software set detects and logs those errors.

The functionality of the memory error handling software can be divided into the following functionality:

• Detailed memory logging:  The software provides a detailed description of a system's DIMM configuration. It tracks DIMMs by serial number, and maintains a separate history of corrected errors for failure analysis. Each log entry includes comprehensive information for each corrected memory error, including rank, bank, row and column. A record of the failure history is maintained for each DIMM and is used to determine if a DIMM failure is a result of a single storage cell or a more widespread problem. Should a pattern of failures emerge, the software issues an alert that the DIMM has reached a critical threshold and should be replaced.
• Predictive failure analysis with proactive actions:  One of the key elements of the memory error handling software is a proprietary transient error filter that identifies and resolves memory problems without triggering downtime. By analyzing logged memory data, the software is able to distinguish correctable transient errors from correctable hard failures and take action accordingly.

When the software detects a persistent DIMM failure, it passes the memory error address to the kernel's page retirement interface to remove the affected page from use. Data is moved from the failing location to a good memory location without interrupting any running jobs, or requiring a system shutdown. Because 4K memory pages are re-mapped at a logical level, the re-mapping is non-disruptive to running jobs.

• Alerts to schedule service:  The memory error handling software sends alerts to HPE Pointnext (when Remote Services is enabled) so any necessary actions can be scheduled. If Remote Services is disabled, notifications are sent to the system administrator.

Power Resource Management

Power data is collected on a per second basis without any latency impact on the system. Up to three months of power data is stored in a database for power utilization behavior analysis that can drive power resource efficiencies for the data center.

Power can be limited on a per-node basis for all nodes or for a specific set of nodes in a system.  This enables finer-grained control over the power usage across the system.

HPE SGI Management Suite supports system-level power management, which allows system administrators to measure and control power and thermals for the entire cluster. Capabilities include the following:

• System power monitoring:  System administrators can read the power and energy, as well as set or read a power limit on the entire cluster, a set of racks, servers or processors.
• System thermal monitoring:  Thermal monitoring involves measuring system temperatures at various points and fan rates. As with power monitoring, systems administrators can specify if they want to view information at the cluster level or rack level.
• System power limiting:  At the system level, users can broadly set a power limit for a rack, set of racks or all racks in a system. For system-level and job-level power limiting, the power managemement software uses the out-of-band node manager capability along with the out-of-band HPE SGI Management Suite power service.  

Power limiting at the job level or resource manager level also uses the node manager capabilities. The HPE SGI Management Suite power service performs power limiting for the resource manager. The node manager's design allows for multiple power policies to be deployed at a single time. In addition, the node manager always chooses the most restrictive power limit of all active policies to engage.

Operating System Compatibility

HPE SGI Management Suite is supported on Red Hat® Enterprise Linux® (RHEL) Advanced Server 6, 7 (through release 7.4), SUSE Linux Enterprise Server (SLES) 11, 12 (through SP3) and CentOS 7.4

NOTE: CentOS, Red Hat^® or a SUSE Linux^® operating system version must be installed on the management server prior to installing HPE SGI Management Suite. This software can be purchased from Hewlett Packard Enterprise and factory installed for the HPE SGI 8600 and Apollo 40 systems.  Please contact your Hewlett Packard Enterprise sales representative for more information.

Hardware Compatibility

HPE SGI Management Suite is supported on the HPE SGI 8600.  Factory installation is available.

HPE SGI Management Suite is also supported on the Apollo 40 system, but only when used in a cluster containing Apollo 40 systems, HPE SGI 8600, or classic SGI Rackable servers.  If used in a cluster with DL servers or other HPE Apollo systems, please order the HPE Insight Cluster Management Utility.

Please contact your Hewlett Packard Enterprise sales representative for more information.

Software Licensing Information

For the Software to be valid on an HPE cluster, each socket in the HPE cluster must have a valid HPE SGI Management Suite license. Subject to the terms and conditions of this Agreement and the payment of any applicable license fee, HPE grants a non-exclusive, non-transferable license to use (as defined below), in object code form, one copy of the Software on one device (socket) at a time for internal business purposes, unless otherwise indicated above or in applicable Transaction Document(s). "Use" means to install, store, load, execute and display the Software in accordance with the Specifications. Use of the Software is subject to these license terms and to the other restrictions specified by Hewlett Packard Enterprise in any other tangible or electronic documentation delivered or otherwise made available with or at the time of purchase of the Software, including license terms, warranty statements, Specifications, and "readme" or other informational files included in the Software itself. Such restrictions are hereby incorporated in this Agreement by reference. Some Software may require license keys or contain other technical protection measures. HPE reserves the right to monitor compliance with Use restrictions remotely or otherwise. Hewlett Packard Enterprise may make a license management program available which records and reports license usage information, If so supplied, customer agrees to install and run such license management program beginning no later than one hundred and eighty (180) days from the date it is made available and continuing for the period that the Software is Used.

Other terms of the HPE Software License are provided on the license agreement that is delivered with the HPE SGI Management Suite software.

Warranty

Hewlett Packard Enterprise will replace defective delivery media for a period of 90 days from the date of purchase. This warranty applies to all HPE SGI Management Suite products found on the delivery media.

HPE Software Support

HPE Pointnext leverages our breadth and depth of technical expertise and innovation to help accelerate digital transformation with Advisory, Professional, and Operational Services. There is a full range of services to complement HPE Performance Software - Message Passing Interface (MPI) software from advisory and design, benchmarking and tuning services, factory pre-installation, configuration, and acceptance as well as training and operational services.

Advisory Services includes design, strategy, road map, and other services to help enable the digital transformation journey, tuned to IT and business needs. Advisory Services helps customers on their journey to Hybrid IT, Big Data, and the Intelligent Edge.

Professional Services helps integrate the new solution with project management, installation and startup, relocation services, and more. In addition, Factory Express installs the software in the factory when building the system. HPE Education Services helps train staff using and managing the software and other technology. We help mitigate risk to the business, so there is no interruption when new technology is being integrated into the existing IT environment. 

Operational Services

• HPE Flexible Capacity is a new consumption model to manage on-demand capacity, combining the agility and economics of public cloud with the security and performance of on-premises IT.
• HPE Datacenter Care offers a tailored operational support solution built on core deliverables. It includes hardware and software support, a team of experts to help personalize deliverables and share best practices, as well as optional building blocks to address specific IT and business needs. HPE Datacenter Care for Hyperscale gives customers access to the Hyperscale Center of Excellence with technical experts who understand how to manage IT at scale including the software.
• HPE Proactive Care is an integrated set of hardware and software support including an enhanced call experience with start to finish case management helping resolve incidents quickly and keeping IT reliable and stable.
• HPE Foundation Care helps when there is a hardware or software problem offering several response levels dependent on IT and business requirements.

Join the Conversation
The HPE Support Center is a community-based, user-supported tool for Hewlett Packard Enterprise customers to participate in discussions amongst the customer community about Hewlett Packard Enterprise products.

http://www.hpe.com/go/hpsc

Contact Support
http://www.hpe.com/services

Date

Version History

Action

Description of Change

4-Dec-2017

Version 1

Created

Created QuickSpecs for HPE SGI Management Suite