Intel E7-4850 v2 CM8063601272906 Benutzerhandbuch
Produktcode
CM8063601272906
Intel
®
Xeon
®
Processor E7-2800/4800/8800 v2 Product Family
51
Datasheet Volume Two: Functional Description, February 2014
Reliability, Availability, Serviceability, and Manageability
7
Reliability, Availability,
Serviceability, and
Manageability
This chapter describes RASM (Reliability, Availability, Serviceability and Manageability)
features of the Intel Xeon processor E7 v2 product family. RASM refers to feature sets
that are associated with system robustness and is defined as follows:
features of the Intel Xeon processor E7 v2 product family. RASM refers to feature sets
that are associated with system robustness and is defined as follows:
Reliability: System capability to detect errors, report errors, and correct errors.
Reliability is typically measured in FITs (Failure in Time). 1 FIT = 1 error in 1 Billion
Hours. Alternatively it can also be measured as MTBF (Mean Time Between Failure).
1FIT is equivalent to approximately 114K Years of MTBF!.
Reliability is typically measured in FITs (Failure in Time). 1 FIT = 1 error in 1 Billion
Hours. Alternatively it can also be measured as MTBF (Mean Time Between Failure).
1FIT is equivalent to approximately 114K Years of MTBF!.
Availability: System capability to redistribute (or reallocate) resources and maintain
normal operation if an error event occurs. Availability is typically measured in “Uptime
or downtime” over a given time interval, for example, 99.999% availability (also known
as ‘Five 9’s). A “Five 9’s Available” system would assure an average “down time” of 5
seconds over one year of continuous operation.
normal operation if an error event occurs. Availability is typically measured in “Uptime
or downtime” over a given time interval, for example, 99.999% availability (also known
as ‘Five 9’s). A “Five 9’s Available” system would assure an average “down time” of 5
seconds over one year of continuous operation.
Serviceability: System capability to effectively report a failure and to expedite the
servicing efforts. Incorporating serviceability features typically results in more efficient
product maintenance, reduces operational costs, and results in minimizing the
“downtime”.
servicing efforts. Incorporating serviceability features typically results in more efficient
product maintenance, reduces operational costs, and results in minimizing the
“downtime”.
Manageability: System capability to monitor the health of the individual components,
to predict failures, and to manage various resources. Incorporating manageability
features typically results in maximizing the ‘uptime’ of a system.
to predict failures, and to manage various resources. Incorporating manageability
features typically results in maximizing the ‘uptime’ of a system.
The primary objective of this chapter is to cover the Reliability features of the processor
since Availability, Serviceability, and Manageability features are generally applicable at
system level. Any feature that also addresses the Availability, Serviceability, and
Manageability will be clarified where applicable.
since Availability, Serviceability, and Manageability features are generally applicable at
system level. Any feature that also addresses the Availability, Serviceability, and
Manageability will be clarified where applicable.
7.1
RASM Overview
for a high level
view of the processor’s key modules. Intel Xeon processor E7 v2 product family
incorporates several features to address system reliability, availability, serviceability,
and manageability requirements. In order to meet the target reliability requirements
and to minimize the impact of the errors, processor incorporates various techniques
such as parity, ECC, CRC, and redundancy within individual modules. Most of these
errors are corrected by the built-in error correction logic and are called as “Corrected
Errors”. Intel Xeon processor E7 v2 product family also incorporates Corrupt Data
Containment and MCA Recovery features to minimize the impact due to certain errors
known as ‘Uncorrected Recoverable” (UCR) errors. This section first describes the
sources of errors, classifies the errors based upon processor’s error handling capability,
and finally briefly documents all the available RASM features. Subsequent sections
describe these RASM features in more detail.
incorporates several features to address system reliability, availability, serviceability,
and manageability requirements. In order to meet the target reliability requirements
and to minimize the impact of the errors, processor incorporates various techniques
such as parity, ECC, CRC, and redundancy within individual modules. Most of these
errors are corrected by the built-in error correction logic and are called as “Corrected
Errors”. Intel Xeon processor E7 v2 product family also incorporates Corrupt Data
Containment and MCA Recovery features to minimize the impact due to certain errors
known as ‘Uncorrected Recoverable” (UCR) errors. This section first describes the
sources of errors, classifies the errors based upon processor’s error handling capability,
and finally briefly documents all the available RASM features. Subsequent sections
describe these RASM features in more detail.