Closed Loop Platform Automation - Tong Zhong & Emma Collins

L
closedloopautomation
Telemetry/Analytics-ml/Orchestration
Q3 2019
Emma Collins/Tong Zhang
2
Legal Disclaimer
General Disclaimer:
© Copyright 2019 Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Inside, the Intel Inside logo, Intel.
Experience What’s Inside are trademarks of Intel. Corporation in the U.S. and/or other countries. *Other names and
brands may be claimed as the property of others.
Technology Disclaimer:
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software
or service activation. Performance varies depending on system configuration. No computer system can be absolutely
secure. Check with your system manufacturer or retailer or learn more at [intel.com].
Performance Disclaimers:
Cost reduction scenarios described are intended as examples of how a given Intel- based product, in the specified
circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does
not guarantee any costs or cost reduction.
Results have been estimated or simulated using internal Intel analysis or architecture simulation or modeling, and
provided to you for informational purposes. Any differences in your system hardware, software or configuration may
affect your actual performance.
ScaleEfficiencywithData-Driven,ClosedLoopAutomation
Intel Platform Features are part of intelligent, closed loop solutions
that are reactive, proactive and predictive, delivering new levels of
efficiency for IT and network infrastructure.
Automated Action Telemetry Analysis
Software and Services
Telemetry
IA Platform Telemetry
Fine-grained Hardware and software insights feeding operational intelligence and automation
Intel's Ingredients for Closed Loop Automation
 Intel driven Analytics solutions using
IA feature data
OR
 Integrate with proprietary/commercial
network monitoring and analytics
solutionsOrchestration
IA Platform
Telemetry
Analytics
 Intel driven MANO solutions to
provide scale/heal/placement
actions that incorporate IA features
OR
 Integrate with
proprietary/commercial MANO
solutions
IA Feature Metrics and
Events Exposure
IA Feature Detection and
Provisioning
Power PMU RDT RAS Other…
The Closed Loop
Intel Platform
Inteltelemetrycollectionandpublication
Platform
Compute Networking Memory Storage Acceleration
Collectd South Bound Plugins
Collectd
Collectd North Bound Plugins
Open stack
MANO
Platform/
NFVI
Monitoring /Analytics
Systems
Telemetry Publication
Telemetry Consumption
Kubernetes
ONAP
Telemetry Collection
Application
6
IntelTelemetryCoverage
Compute Network Storage
Hypervisor
NFVIVirtualised
Compute
Virtualised
Network
Virtualised
Storage
Collectd
PMU
counters
NIC counters
vSwitch
counters
Common / Standard Open APIs
VM Stall
Detection/
RT Stall Detection
Enterprise and Network Management Tools
RAS
Hypervisor/Container
Counters
Intel® Node Manager
Open Platform
Collector
Intel® Run Sure Technology
MCA* PCIe AER
Resilient System Technology
Resilient Memory Technology
SDDC DDDC+1 Mirroring
RAID
Intel® Rapid
Storage
Technology
Intel®
Management
Engine
IPMI
C
M
T
Intel® RDT
C
A
T
M
B
M
C
D
P
P
O
W
E
R
VIM
Intel® Infrastructure
Management
Technologies
Redfish SYSLOG KafkaSNMP API VES Plugin Prometheus
OpenStack Kubernetes
TyingtheIntelPlatformtotheCustomerBusinessCase
Customer Business Use Case
Platform
Slicing
(Part of 5G Network
Slicing)
Platform
Resiliency
Security
(side channel attacks)
Power
Optimisation
Intel Platform
Power PMU RDT RAS Other…
MANO
Analytics
MachineLearningWillCompletetheEvolution
Learn
Learn
Learn
Learn
Watch Decide
Collect
Act
App
Server/Storage/Network
Watch
Expose Platform and
App data through
APIs
Act
Mechanism for
Policy Activation
and Enforcement
Learn
Machine Learning
for continuous
improvement
Decide
Analytics to define
correct response
Moving From Automation to Self-OptimizationTaking the manpower out of networking, cloud…
DeliveringClosedLoopAutomationforNFV
Automated
Action
IA Platform
Telemetry
Telemetry
Analysis
1. Enable IA Service Assurance Telemetry
through instrumentation and exposure in
an industry standard manner
2. Enable IA Telemetry Analytics through
Telemetry Compaction, KPI identification
and prediction
3. Enable Closed Loop Automation through
Orchestration enabling and industry
proof points
Casestudy:MLforNFVserviceassurance
10
• Efficient dynamic network management is one major challenge for NFV
• Machine learning plays an important role in addressing this challenge by
analyzing gathered data for various purposes:
• dynamic resource allocation
• security threats alert
• performance degradation detection
• demand prediction
“Cognet: A Network Management Architecture Featuring Cognitive Capabilities,” Proc. Euro. Conf. Networks and Commun., June 2016
1. Data pre-processing
Feature Engineering/Reduction
2. KPI prediction/forecasting
Regression/Classification
3. Closed loop optimization
Reinforcement Learning
NFVtestsystems
11
vEPC – virtual Evolved Packet Core
vCMTS – virtual Cable Modem Termination System
• Telemetry data dumped through Collectd includes CPU, PMU, Memory, Load, etc.
• Total number of telemetry data hundreds to thousands sampled at configured interval (1s or 10s)
• Target KPI (Key Performance Indicator): packet drop rate
Datapre-processing
12
• Data filtering – remove irrelevant data (e.g. control plane data for this case)
 1065 features remaining
• Data alignment and interpolation
• Feature selection – remove features with no change over time
 726 features remaining
• Data normalization
• Data splitting to training, validation and testing sets
 Tens of thousands of samples split into 8:1:1
Telemetryfeaturecompaction
• Feature Selection
• The process of selecting a subset of relevant features for use in the model construction
• Filter methods – Select features based on scoring from statistical measures e.g. SelectKBest
• Wrapper methods – search for optimal feature combination that results in best predictive results
e.g. Recursive Feature Elimination, Boruta
• Unsupervised learning methods – group features that behave similarly e.g. FeatureAgglomeration
• Feature Transformation (dimension reduction)
• Convert the feature vector into lower dimension space with learned transformations
• Supervised: PLS (Partial Least Squares), CCA (Canonical Correlation Analysis), LDA (Linear
Discriminant Analysis)
• Unsupervised: PCA (Principal Component Analysis)
13
10 20 40 Feature 10 20 40 Feature 10 20 40 Feature
• cpu_value_idle_18 • • intel_rdt_value_bytes_llc_20 • • • intel_rdt_value_memory_bandwidth_local_12
• cpu_value_idle_49 • • intel_rdt_value_bytes_llc_27 • intel_rdt_value_memory_bandwidth_local_3
• • cpu_value_interrupt_18 • intel_rdt_value_bytes_llc_41 • intel_rdt_value_memory_bandwidth_local_33
• cpu_value_interrupt_20 • • intel_rdt_value_bytes_llc_50 • intel_rdt_value_memory_bandwidth_local_53
• cpu_value_system_39 • intel_rdt_value_bytes_llc_53 • • intel_rdt_value_memory_bandwidth_local_9
• • cpu_value_user_13 • • intel_rdt_value_bytes_llc_6 • intel_rdt_value_memory_bandwidth_remote_2
• cpu_value_user_3 • intel_rdt_value_bytes_llc_8 • intel_rdt_value_memory_bandwidth_remote_6
• cpu_value_user_46 • intel_rdt_value_ipc_nan_22 • • ipmi_value_fanspeed_System Fan 1 fan_cooling (29.1)
• cpu_value_user_9 • • intel_rdt_value_ipc_nan_23 • ipmi_value_temperature_HSBP 1 Temp drive_backplane (15.1)
• df_value_free_etc-hosts • intel_rdt_value_ipc_nan_26 • ipmi_value_temperature_LAN NICTemp system_board (7.1)
• df_value_used_etc-hosts • intel_rdt_value_ipc_nan_5 • ipmi_value_temperature_P2 DTS Therm Mgn processor (3.2)
• disk_read_disk_time_sda • • intel_rdt_value_ipc_nan_50 • • • irq_value_TLB
• intel_pmu_value_branches_42 • intel_rdt_value_ipc_nan_53 • • • load_longterm
• intel_pmu_value_instructions_14 • intel_rdt_value_ipc_nan_54 • • • load_midterm
• intel_pmu_value_page-faults_51 • intel_rdt_value_ipc_nan_9 • load_shortterm
• intel_pmu_value_page-faults_all • numa_value_other_node_node1 • • • memory_value_cached
• • memory_value_slab_unrecl
Top TopTop
TopTelemetryFeaturesSelectedByML
14
Recursive feature elimination stepping down the target from 40  20  10
Featureselectionresults
15
• Applying various feature selection algorithms to reduce the number of telemetry data sampled without
compromising the prediction accuracy
• Smaller set of telemetry data saves training/inference time
• vEPC test data set with original feature dimension ~400
• GradientBoostRegressor algorithm used for packet loss rate prediction
• Select top number of features from feature importance output of GradientBoostRegressor
KPIprediction&forecasting–supervisedlearning
16
• Regression
• Train the model to predict target KPI using reduced telemetry data samples
• Example: Packet drop rate
• Accuracy Measured by MSE (Mean Squared Error) – the smaller the better
• Classification
• Detect packet drop from telemetry data
• Measurement
• Accuracy_score = correct_prediction / total_sample
• Precision/Recall/F1-score, etc.
• Time Series Forecasting
• Predict future KPI value based on historical telemetry data and observed KPI trend
KPIforecastingusingLSTM
17
• vCMTS downlink test data
• KPI – scheduling packet loss rate
• Input: previous 60 seconds of 20 selected
telemetry data + KPI
• Output: predicted future value (5 seconds
later) of target KPI
• Tensorflow BasicLSTMCell: Two layers,
each layer 150 neurons
Closed-loopAutomation Xeon
vCMTS -
0
Traffic Generator
vCMTS -
0VNFs
NIC
Machine Learning
Modules
collectd
InfluxDB
Traffic
forecasting
Reinforcement
learning
HW
Optimization
(RDT, DVFS, WL
consolidation, etc.)
Compacted
metrics
Using ML:
• Track/forecast workload,
performance
• Dynamically adjust
resource allocation
Benefits:
• Reduced TCO through
power saving and
increased HW utilization
Frequencytuning
ENIPoC–NetworkSliceLifecycleManagement
20
• For generating new scale
up/down and converting the
intent to suggested configuration
• LSTM is used for traffic prediction
AI-based predictor:
TNSM:
CNSM:
• Provides underlay network
control to satisfy the network
slice requests
• FlexE and a FlexE-based
optimization algorithm are used
for underlay network slice
creation and modification
• Provides core network control
to satisfy the network slice
requests
ETSI ISG ENI – Experiential Networked Intelligence
Trafficpredictionforresourceoptimization
21
Blue: actual traffic
Orange: predicted traffic
AIsystemarchitecture
22
Source:
Intel Confidential
Backup
23
Further Resources
Learn more from these helpful sites:
https://networkbuilders.intel.com/network-technologies/serviceassurance
https://wiki.opnfv.org/display/fastpath/Barometer+Home
https://wiki.openstack.org/wiki/Telemetry
https://01.org/openstack/blogs/2015/openstack-enhanced-platform-
awareness-white-paper
25
Collectd101materials
• Collectd 101
• https://wiki.opnfv.org/display/fastpath/Collectd+101
• Write simple read plugin
• https://wiki.opnfv.org/display/fastpath/Collectd+how+to+implement+a+si
mple+plugin
26
Barometer Strategy:
• Ensure platform metrics/events are accessible
through open industry standard interfaces.
• Demonstrate IA platform technologies can be
monitored, consumed and actioned in real time
Opnfvbarometer–Intelplatformfeatureplugins
One Click Install:
• Easy install/configuration
for customers
• One command to install
Collectd/Influxdb/Grafana
Three container approach for Collectd:
• Stable Container: latest stable branch
• Master Container: up to date with master
• Experimental Container: cherry pick
features of interest
BarometerLinks
Barometer Home: https://wiki.opnfv.org/display/fastpath/Barometer+Home
Metrics/Events through Barometer (not on Collectd site):
https://wiki.opnfv.org/display/fastpath/Collectd+Metrics+and+Events#CollectdM
etricsandEvents-Metrics
Barometer “One-click” install:
https://wiki.opnfv.org/display/fastpath/One+Click+Install+of+Barometer+Contain
ers
27
1 von 27

Recomendados

Service Assurance Constructs for Achieving Network Transformation - Sunku Ran... von
Service Assurance Constructs for Achieving Network Transformation - Sunku Ran...Service Assurance Constructs for Achieving Network Transformation - Sunku Ran...
Service Assurance Constructs for Achieving Network Transformation - Sunku Ran...Liz Warner
120 views44 Folien
Closed Loop Network Automation for Optimal Resource Allocation via Reinforcem... von
Closed Loop Network Automation for Optimal Resource Allocation via Reinforcem...Closed Loop Network Automation for Optimal Resource Allocation via Reinforcem...
Closed Loop Network Automation for Optimal Resource Allocation via Reinforcem...Liz Warner
114 views23 Folien
Platform Observability and Infrastructure Closed Loops von
Platform Observability and Infrastructure Closed LoopsPlatform Observability and Infrastructure Closed Loops
Platform Observability and Infrastructure Closed LoopsLiz Warner
60 views19 Folien
Im 2021 tutorial next-generation closed-loop automation - an inside view - ... von
Im 2021 tutorial   next-generation closed-loop automation - an inside view - ...Im 2021 tutorial   next-generation closed-loop automation - an inside view - ...
Im 2021 tutorial next-generation closed-loop automation - an inside view - ...Ishan Vaishnavi
49 views122 Folien
Intel® Select Solutions for the Network von
Intel® Select Solutions for the NetworkIntel® Select Solutions for the Network
Intel® Select Solutions for the NetworkLiz Warner
279 views25 Folien
Running Kubernetes on OpenStack von
Running Kubernetes on OpenStackRunning Kubernetes on OpenStack
Running Kubernetes on OpenStackLiz Warner
93 views19 Folien

Más contenido relacionado

Was ist angesagt?

Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors von
Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors
Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors Michelle Holley
1.4K views24 Folien
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me... von
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...Haidee McMahon
1.1K views15 Folien
Development, test, and characterization of MEC platforms with Teranium and Dr... von
Development, test, and characterization of MEC platforms with Teranium and Dr...Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...Michelle Holley
330 views24 Folien
Using Xeon + FPGA for Accelerating HPC Workloads von
Using Xeon + FPGA for Accelerating HPC WorkloadsUsing Xeon + FPGA for Accelerating HPC Workloads
Using Xeon + FPGA for Accelerating HPC Workloadsinside-BigData.com
4.5K views35 Folien
Edge and 5G: What is in it for the developers? von
Edge and 5G: What is in it for the developers?Edge and 5G: What is in it for the developers?
Edge and 5G: What is in it for the developers?Michelle Holley
1.5K views17 Folien
FPGAs and Machine Learning von
FPGAs and Machine LearningFPGAs and Machine Learning
FPGAs and Machine Learninginside-BigData.com
4.4K views179 Folien

Was ist angesagt?(18)

Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors von Michelle Holley
Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors
Building efficient 5G NR base stations with Intel® Xeon® Scalable Processors
Michelle Holley1.4K views
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me... von Haidee McMahon
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...
Ligato - A platform for development of Cloud-Native VNF's - SDN/NFV London me...
Haidee McMahon1.1K views
Development, test, and characterization of MEC platforms with Teranium and Dr... von Michelle Holley
Development, test, and characterization of MEC platforms with Teranium and Dr...Development, test, and characterization of MEC platforms with Teranium and Dr...
Development, test, and characterization of MEC platforms with Teranium and Dr...
Michelle Holley330 views
Using Xeon + FPGA for Accelerating HPC Workloads von inside-BigData.com
Using Xeon + FPGA for Accelerating HPC WorkloadsUsing Xeon + FPGA for Accelerating HPC Workloads
Using Xeon + FPGA for Accelerating HPC Workloads
inside-BigData.com4.5K views
Edge and 5G: What is in it for the developers? von Michelle Holley
Edge and 5G: What is in it for the developers?Edge and 5G: What is in it for the developers?
Edge and 5G: What is in it for the developers?
Michelle Holley1.5K views
Distributed Resource Management Application API (DRMAA) Version 2 von Peter Tröger
Distributed Resource Management Application API (DRMAA) Version 2Distributed Resource Management Application API (DRMAA) Version 2
Distributed Resource Management Application API (DRMAA) Version 2
Peter Tröger283 views
Task allocation on many core-multi processor distributed system von Deepak Shankar
Task allocation on many core-multi processor distributed systemTask allocation on many core-multi processor distributed system
Task allocation on many core-multi processor distributed system
Deepak Shankar90 views
Introduction to architecture exploration von Deepak Shankar
Introduction to architecture explorationIntroduction to architecture exploration
Introduction to architecture exploration
Deepak Shankar139 views
Preparing to program Aurora at Exascale - Early experiences and future direct... von inside-BigData.com
Preparing to program Aurora at Exascale - Early experiences and future direct...Preparing to program Aurora at Exascale - Early experiences and future direct...
Preparing to program Aurora at Exascale - Early experiences and future direct...
inside-BigData.com1.4K views
Exploration of Radars and Software Defined Radios using VisualSim von Deepak Shankar
Exploration of  Radars and Software Defined Radios using VisualSimExploration of  Radars and Software Defined Radios using VisualSim
Exploration of Radars and Software Defined Radios using VisualSim
Deepak Shankar45 views
Centralized Emergency Traffic Optimizer NEV SDK von Michelle Holley
Centralized Emergency Traffic Optimizer NEV SDKCentralized Emergency Traffic Optimizer NEV SDK
Centralized Emergency Traffic Optimizer NEV SDK
Michelle Holley305 views
Comparative Analysis of IT Monitoring Tools von apprize360
Comparative Analysis of IT Monitoring ToolsComparative Analysis of IT Monitoring Tools
Comparative Analysis of IT Monitoring Tools
apprize3607.4K views
Data Center Design Guide 4 2 von Fiyaz Syed
Data Center Design Guide 4 2Data Center Design Guide 4 2
Data Center Design Guide 4 2
Fiyaz Syed1.7K views
Storage Networking Solutions for High Performance Databases by QLogic von Jone Smith
Storage Networking Solutions for High Performance Databases by QLogicStorage Networking Solutions for High Performance Databases by QLogic
Storage Networking Solutions for High Performance Databases by QLogic
Jone Smith511 views
ETSI NFV#13 NFV resiliency presentation - ali kafel - stratus von Ali Kafel
ETSI NFV#13   NFV resiliency presentation - ali kafel - stratusETSI NFV#13   NFV resiliency presentation - ali kafel - stratus
ETSI NFV#13 NFV resiliency presentation - ali kafel - stratus
Ali Kafel677 views

Similar a Closed Loop Platform Automation - Tong Zhong & Emma Collins

Performance out of the box developers von
Performance   out of the box developersPerformance   out of the box developers
Performance out of the box developersMichelle Holley
313 views14 Folien
Strata + Hadoop 2015 Slides von
Strata + Hadoop 2015 SlidesStrata + Hadoop 2015 Slides
Strata + Hadoop 2015 SlidesJun Liu
258 views44 Folien
Design and Optimize your code for high-performance with Intel® Advisor and I... von
Design and Optimize your code for high-performance with Intel®  Advisor and I...Design and Optimize your code for high-performance with Intel®  Advisor and I...
Design and Optimize your code for high-performance with Intel® Advisor and I...Tyrone Systems
208 views30 Folien
Yashi dealer meeting settembre 2016 tecnologie xeon intel italia von
Yashi dealer meeting settembre 2016 tecnologie xeon intel italiaYashi dealer meeting settembre 2016 tecnologie xeon intel italia
Yashi dealer meeting settembre 2016 tecnologie xeon intel italiaYashi Italia
170 views16 Folien
6 profiling tools von
6 profiling tools6 profiling tools
6 profiling toolsvideos
3.5K views48 Folien
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra... von
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...Liz Warner
212 views44 Folien

Similar a Closed Loop Platform Automation - Tong Zhong & Emma Collins(20)

Performance out of the box developers von Michelle Holley
Performance   out of the box developersPerformance   out of the box developers
Performance out of the box developers
Michelle Holley313 views
Strata + Hadoop 2015 Slides von Jun Liu
Strata + Hadoop 2015 SlidesStrata + Hadoop 2015 Slides
Strata + Hadoop 2015 Slides
Jun Liu258 views
Design and Optimize your code for high-performance with Intel® Advisor and I... von Tyrone Systems
Design and Optimize your code for high-performance with Intel®  Advisor and I...Design and Optimize your code for high-performance with Intel®  Advisor and I...
Design and Optimize your code for high-performance with Intel® Advisor and I...
Tyrone Systems208 views
Yashi dealer meeting settembre 2016 tecnologie xeon intel italia von Yashi Italia
Yashi dealer meeting settembre 2016 tecnologie xeon intel italiaYashi dealer meeting settembre 2016 tecnologie xeon intel italia
Yashi dealer meeting settembre 2016 tecnologie xeon intel italia
Yashi Italia170 views
6 profiling tools von videos
6 profiling tools6 profiling tools
6 profiling tools
videos3.5K views
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra... von Liz Warner
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...
Service Assurance Constructs for Achieving Network Transformation by Sunku Ra...
Liz Warner212 views
MCSoC'13 Keynote Talk "Taming Big Data Streams" von Hideyuki Kawashima
MCSoC'13 Keynote Talk "Taming Big Data Streams"MCSoC'13 Keynote Talk "Taming Big Data Streams"
MCSoC'13 Keynote Talk "Taming Big Data Streams"
Hideyuki Kawashima824 views
Performance and Power Profiling on Intel Android Devices von Intel® Software
Performance and Power Profiling on Intel Android DevicesPerformance and Power Profiling on Intel Android Devices
Performance and Power Profiling on Intel Android Devices
Intel® Software1.3K views
Web Crawler Detection Model von MatinZivdar
Web Crawler Detection ModelWeb Crawler Detection Model
Web Crawler Detection Model
MatinZivdar70 views
Netflix SRE perf meetup_slides von Ed Hunter
Netflix SRE perf meetup_slidesNetflix SRE perf meetup_slides
Netflix SRE perf meetup_slides
Ed Hunter3K views
P4/FPGA, Packet Acceleration von Liz Warner
P4/FPGA, Packet AccelerationP4/FPGA, Packet Acceleration
P4/FPGA, Packet Acceleration
Liz Warner510 views
Basics of embedded system design von K Senthil Kumar
Basics of embedded system designBasics of embedded system design
Basics of embedded system design
K Senthil Kumar2.1K views
Extend HPC Workloads to Amazon EC2 Instances with Intel and Rescale (CMP373-S... von Amazon Web Services
Extend HPC Workloads to Amazon EC2 Instances with Intel and Rescale (CMP373-S...Extend HPC Workloads to Amazon EC2 Instances with Intel and Rescale (CMP373-S...
Extend HPC Workloads to Amazon EC2 Instances with Intel and Rescale (CMP373-S...
MySQL-Performance Schema- What's new in MySQL-5.7 DMRs von Mayank Prasad
MySQL-Performance Schema- What's new in MySQL-5.7 DMRsMySQL-Performance Schema- What's new in MySQL-5.7 DMRs
MySQL-Performance Schema- What's new in MySQL-5.7 DMRs
Mayank Prasad951 views
Enabling Applications to Exploit SmartNICs and FPGAs von inside-BigData.com
Enabling Applications to Exploit SmartNICs and FPGAsEnabling Applications to Exploit SmartNICs and FPGAs
Enabling Applications to Exploit SmartNICs and FPGAs
inside-BigData.com670 views

Más de Liz Warner

Open Source 5G/Edge Automation via ONAP von
Open Source 5G/Edge Automation via ONAPOpen Source 5G/Edge Automation via ONAP
Open Source 5G/Edge Automation via ONAPLiz Warner
663 views20 Folien
Easing the Path to Network Transformation - Network Transformation Experience... von
Easing the Path to Network Transformation - Network Transformation Experience...Easing the Path to Network Transformation - Network Transformation Experience...
Easing the Path to Network Transformation - Network Transformation Experience...Liz Warner
175 views38 Folien
CNTT with Airship von
CNTT with AirshipCNTT with Airship
CNTT with AirshipLiz Warner
138 views14 Folien
Your Path to Edge Computing - Akraino Edge Stack Update von
Your Path to Edge Computing - Akraino Edge Stack UpdateYour Path to Edge Computing - Akraino Edge Stack Update
Your Path to Edge Computing - Akraino Edge Stack UpdateLiz Warner
443 views11 Folien
Introduction to Tungsten Fabric and the vRouter von
Introduction to Tungsten Fabric and the vRouterIntroduction to Tungsten Fabric and the vRouter
Introduction to Tungsten Fabric and the vRouterLiz Warner
587 views16 Folien
Linux Akraino Blueprint von
Linux Akraino BlueprintLinux Akraino Blueprint
Linux Akraino BlueprintLiz Warner
209 views13 Folien

Más de Liz Warner(17)

Open Source 5G/Edge Automation via ONAP von Liz Warner
Open Source 5G/Edge Automation via ONAPOpen Source 5G/Edge Automation via ONAP
Open Source 5G/Edge Automation via ONAP
Liz Warner663 views
Easing the Path to Network Transformation - Network Transformation Experience... von Liz Warner
Easing the Path to Network Transformation - Network Transformation Experience...Easing the Path to Network Transformation - Network Transformation Experience...
Easing the Path to Network Transformation - Network Transformation Experience...
Liz Warner175 views
CNTT with Airship von Liz Warner
CNTT with AirshipCNTT with Airship
CNTT with Airship
Liz Warner138 views
Your Path to Edge Computing - Akraino Edge Stack Update von Liz Warner
Your Path to Edge Computing - Akraino Edge Stack UpdateYour Path to Edge Computing - Akraino Edge Stack Update
Your Path to Edge Computing - Akraino Edge Stack Update
Liz Warner443 views
Introduction to Tungsten Fabric and the vRouter von Liz Warner
Introduction to Tungsten Fabric and the vRouterIntroduction to Tungsten Fabric and the vRouter
Introduction to Tungsten Fabric and the vRouter
Liz Warner587 views
Linux Akraino Blueprint von Liz Warner
Linux Akraino BlueprintLinux Akraino Blueprint
Linux Akraino Blueprint
Liz Warner209 views
ONAP and the K8s Ecosystem: A Converged Edge Application & Network Function P... von Liz Warner
ONAP and the K8s Ecosystem: A Converged Edge Application & Network Function P...ONAP and the K8s Ecosystem: A Converged Edge Application & Network Function P...
ONAP and the K8s Ecosystem: A Converged Edge Application & Network Function P...
Liz Warner330 views
Enabling the Deployment of Edge Services with the Open Network Edge Services ... von Liz Warner
Enabling the Deployment of Edge Services with the Open Network Edge Services ...Enabling the Deployment of Edge Services with the Open Network Edge Services ...
Enabling the Deployment of Edge Services with the Open Network Edge Services ...
Liz Warner1.1K views
Unleashing the Power of Fabric Orchestrating New Performance Features for SR-... von Liz Warner
Unleashing the Power of Fabric Orchestrating New Performance Features for SR-...Unleashing the Power of Fabric Orchestrating New Performance Features for SR-...
Unleashing the Power of Fabric Orchestrating New Performance Features for SR-...
Liz Warner109 views
Closed-Loop Platform Automation by Tong Zhong and Emma Collins von Liz Warner
Closed-Loop Platform Automation by Tong Zhong and Emma CollinsClosed-Loop Platform Automation by Tong Zhong and Emma Collins
Closed-Loop Platform Automation by Tong Zhong and Emma Collins
Liz Warner89 views
Closed-Loop Network Automation for Optimal Resource Allocation via Reinforcem... von Liz Warner
Closed-Loop Network Automation for Optimal Resource Allocation via Reinforcem...Closed-Loop Network Automation for Optimal Resource Allocation via Reinforcem...
Closed-Loop Network Automation for Optimal Resource Allocation via Reinforcem...
Liz Warner227 views
Open Network Edge Services Software for 5G and Edge von Liz Warner
Open Network Edge Services Software for 5G and EdgeOpen Network Edge Services Software for 5G and Edge
Open Network Edge Services Software for 5G and Edge
Liz Warner279 views
Akraino and Edge Computing von Liz Warner
Akraino and Edge ComputingAkraino and Edge Computing
Akraino and Edge Computing
Liz Warner475 views
Whats New with Kata Containers von Liz Warner
Whats New with Kata ContainersWhats New with Kata Containers
Whats New with Kata Containers
Liz Warner97 views
SEBA: SDN Enabled Broadband Access - Transporting SDN principles to PON Networks von Liz Warner
SEBA: SDN Enabled Broadband Access - Transporting SDN principles to PON NetworksSEBA: SDN Enabled Broadband Access - Transporting SDN principles to PON Networks
SEBA: SDN Enabled Broadband Access - Transporting SDN principles to PON Networks
Liz Warner516 views
Simplifying and accelerating converged media with Open Visual Cloud von Liz Warner
Simplifying and accelerating converged media with Open Visual CloudSimplifying and accelerating converged media with Open Visual Cloud
Simplifying and accelerating converged media with Open Visual Cloud
Liz Warner318 views
Open Source for the 4th Industrial Revolution von Liz Warner
Open Source for the 4th Industrial RevolutionOpen Source for the 4th Industrial Revolution
Open Source for the 4th Industrial Revolution
Liz Warner204 views

Último

The_Chronological_Life_of_Christ_Part_82_I_Amx von
The_Chronological_Life_of_Christ_Part_82_I_AmxThe_Chronological_Life_of_Christ_Part_82_I_Amx
The_Chronological_Life_of_Christ_Part_82_I_AmxNetwork Bible Fellowship
7 views17 Folien
Spiritual Journey by Gabriel Lopez von
Spiritual Journey by Gabriel LopezSpiritual Journey by Gabriel Lopez
Spiritual Journey by Gabriel Lopez444spiritualmeaning
10 views22 Folien
English - Book of Enoch by R.H. Charles.pdf von
English - Book of Enoch by R.H. Charles.pdfEnglish - Book of Enoch by R.H. Charles.pdf
English - Book of Enoch by R.H. Charles.pdfFilipino Tracts and Literature Society Inc.
5 views32 Folien
Bulletin Final.pdf von
Bulletin Final.pdfBulletin Final.pdf
Bulletin Final.pdfbusiness86
32 views4 Folien
Christ the King Sunday von
Christ the King SundayChrist the King Sunday
Christ the King SundayVintage Church
8 views26 Folien
Formed: Finding the Truth About Who You Are von
Formed: Finding the Truth About Who You AreFormed: Finding the Truth About Who You Are
Formed: Finding the Truth About Who You AreCole Hartman
22 views22 Folien

Último(15)

Bulletin Final.pdf von business86
Bulletin Final.pdfBulletin Final.pdf
Bulletin Final.pdf
business8632 views
Formed: Finding the Truth About Who You Are von Cole Hartman
Formed: Finding the Truth About Who You AreFormed: Finding the Truth About Who You Are
Formed: Finding the Truth About Who You Are
Cole Hartman22 views
A331 Don’t you know that you are the temple of God? God's temple, destroy Go... von franktsao4
A331 Don’t you know that you are the temple of God?  God's temple, destroy Go...A331 Don’t you know that you are the temple of God?  God's temple, destroy Go...
A331 Don’t you know that you are the temple of God? God's temple, destroy Go...
franktsao49 views
Homily: Christ The King 2023.docx von James Knipper
Homily: Christ The King 2023.docxHomily: Christ The King 2023.docx
Homily: Christ The King 2023.docx
James Knipper30 views
KingsandProphets2.pptx von Vaughndj
KingsandProphets2.pptxKingsandProphets2.pptx
KingsandProphets2.pptx
Vaughndj11 views
December 2023 Calendar of Events von Floodwoodvern
December 2023 Calendar of EventsDecember 2023 Calendar of Events
December 2023 Calendar of Events
Floodwoodvern8 views
10 LIFE LESSONS ~ The Art of Happiness from the Dalai Lama (Eng & Chi).pptx von OH TEIK BIN
10 LIFE LESSONS ~ The Art of Happiness from the Dalai Lama (Eng & Chi).pptx10 LIFE LESSONS ~ The Art of Happiness from the Dalai Lama (Eng & Chi).pptx
10 LIFE LESSONS ~ The Art of Happiness from the Dalai Lama (Eng & Chi).pptx
OH TEIK BIN23 views

Closed Loop Platform Automation - Tong Zhong & Emma Collins

  • 2. 2 Legal Disclaimer General Disclaimer: © Copyright 2019 Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Inside, the Intel Inside logo, Intel. Experience What’s Inside are trademarks of Intel. Corporation in the U.S. and/or other countries. *Other names and brands may be claimed as the property of others. Technology Disclaimer: Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at [intel.com]. Performance Disclaimers: Cost reduction scenarios described are intended as examples of how a given Intel- based product, in the specified circumstances and configurations, may affect future costs and provide cost savings. Circumstances will vary. Intel does not guarantee any costs or cost reduction. Results have been estimated or simulated using internal Intel analysis or architecture simulation or modeling, and provided to you for informational purposes. Any differences in your system hardware, software or configuration may affect your actual performance.
  • 3. ScaleEfficiencywithData-Driven,ClosedLoopAutomation Intel Platform Features are part of intelligent, closed loop solutions that are reactive, proactive and predictive, delivering new levels of efficiency for IT and network infrastructure. Automated Action Telemetry Analysis Software and Services Telemetry IA Platform Telemetry Fine-grained Hardware and software insights feeding operational intelligence and automation
  • 4. Intel's Ingredients for Closed Loop Automation  Intel driven Analytics solutions using IA feature data OR  Integrate with proprietary/commercial network monitoring and analytics solutionsOrchestration IA Platform Telemetry Analytics  Intel driven MANO solutions to provide scale/heal/placement actions that incorporate IA features OR  Integrate with proprietary/commercial MANO solutions IA Feature Metrics and Events Exposure IA Feature Detection and Provisioning Power PMU RDT RAS Other… The Closed Loop Intel Platform
  • 5. Inteltelemetrycollectionandpublication Platform Compute Networking Memory Storage Acceleration Collectd South Bound Plugins Collectd Collectd North Bound Plugins Open stack MANO Platform/ NFVI Monitoring /Analytics Systems Telemetry Publication Telemetry Consumption Kubernetes ONAP Telemetry Collection Application
  • 6. 6 IntelTelemetryCoverage Compute Network Storage Hypervisor NFVIVirtualised Compute Virtualised Network Virtualised Storage Collectd PMU counters NIC counters vSwitch counters Common / Standard Open APIs VM Stall Detection/ RT Stall Detection Enterprise and Network Management Tools RAS Hypervisor/Container Counters Intel® Node Manager Open Platform Collector Intel® Run Sure Technology MCA* PCIe AER Resilient System Technology Resilient Memory Technology SDDC DDDC+1 Mirroring RAID Intel® Rapid Storage Technology Intel® Management Engine IPMI C M T Intel® RDT C A T M B M C D P P O W E R VIM Intel® Infrastructure Management Technologies Redfish SYSLOG KafkaSNMP API VES Plugin Prometheus OpenStack Kubernetes
  • 7. TyingtheIntelPlatformtotheCustomerBusinessCase Customer Business Use Case Platform Slicing (Part of 5G Network Slicing) Platform Resiliency Security (side channel attacks) Power Optimisation Intel Platform Power PMU RDT RAS Other… MANO Analytics
  • 8. MachineLearningWillCompletetheEvolution Learn Learn Learn Learn Watch Decide Collect Act App Server/Storage/Network Watch Expose Platform and App data through APIs Act Mechanism for Policy Activation and Enforcement Learn Machine Learning for continuous improvement Decide Analytics to define correct response Moving From Automation to Self-OptimizationTaking the manpower out of networking, cloud…
  • 9. DeliveringClosedLoopAutomationforNFV Automated Action IA Platform Telemetry Telemetry Analysis 1. Enable IA Service Assurance Telemetry through instrumentation and exposure in an industry standard manner 2. Enable IA Telemetry Analytics through Telemetry Compaction, KPI identification and prediction 3. Enable Closed Loop Automation through Orchestration enabling and industry proof points
  • 10. Casestudy:MLforNFVserviceassurance 10 • Efficient dynamic network management is one major challenge for NFV • Machine learning plays an important role in addressing this challenge by analyzing gathered data for various purposes: • dynamic resource allocation • security threats alert • performance degradation detection • demand prediction “Cognet: A Network Management Architecture Featuring Cognitive Capabilities,” Proc. Euro. Conf. Networks and Commun., June 2016 1. Data pre-processing Feature Engineering/Reduction 2. KPI prediction/forecasting Regression/Classification 3. Closed loop optimization Reinforcement Learning
  • 11. NFVtestsystems 11 vEPC – virtual Evolved Packet Core vCMTS – virtual Cable Modem Termination System • Telemetry data dumped through Collectd includes CPU, PMU, Memory, Load, etc. • Total number of telemetry data hundreds to thousands sampled at configured interval (1s or 10s) • Target KPI (Key Performance Indicator): packet drop rate
  • 12. Datapre-processing 12 • Data filtering – remove irrelevant data (e.g. control plane data for this case)  1065 features remaining • Data alignment and interpolation • Feature selection – remove features with no change over time  726 features remaining • Data normalization • Data splitting to training, validation and testing sets  Tens of thousands of samples split into 8:1:1
  • 13. Telemetryfeaturecompaction • Feature Selection • The process of selecting a subset of relevant features for use in the model construction • Filter methods – Select features based on scoring from statistical measures e.g. SelectKBest • Wrapper methods – search for optimal feature combination that results in best predictive results e.g. Recursive Feature Elimination, Boruta • Unsupervised learning methods – group features that behave similarly e.g. FeatureAgglomeration • Feature Transformation (dimension reduction) • Convert the feature vector into lower dimension space with learned transformations • Supervised: PLS (Partial Least Squares), CCA (Canonical Correlation Analysis), LDA (Linear Discriminant Analysis) • Unsupervised: PCA (Principal Component Analysis) 13
  • 14. 10 20 40 Feature 10 20 40 Feature 10 20 40 Feature • cpu_value_idle_18 • • intel_rdt_value_bytes_llc_20 • • • intel_rdt_value_memory_bandwidth_local_12 • cpu_value_idle_49 • • intel_rdt_value_bytes_llc_27 • intel_rdt_value_memory_bandwidth_local_3 • • cpu_value_interrupt_18 • intel_rdt_value_bytes_llc_41 • intel_rdt_value_memory_bandwidth_local_33 • cpu_value_interrupt_20 • • intel_rdt_value_bytes_llc_50 • intel_rdt_value_memory_bandwidth_local_53 • cpu_value_system_39 • intel_rdt_value_bytes_llc_53 • • intel_rdt_value_memory_bandwidth_local_9 • • cpu_value_user_13 • • intel_rdt_value_bytes_llc_6 • intel_rdt_value_memory_bandwidth_remote_2 • cpu_value_user_3 • intel_rdt_value_bytes_llc_8 • intel_rdt_value_memory_bandwidth_remote_6 • cpu_value_user_46 • intel_rdt_value_ipc_nan_22 • • ipmi_value_fanspeed_System Fan 1 fan_cooling (29.1) • cpu_value_user_9 • • intel_rdt_value_ipc_nan_23 • ipmi_value_temperature_HSBP 1 Temp drive_backplane (15.1) • df_value_free_etc-hosts • intel_rdt_value_ipc_nan_26 • ipmi_value_temperature_LAN NICTemp system_board (7.1) • df_value_used_etc-hosts • intel_rdt_value_ipc_nan_5 • ipmi_value_temperature_P2 DTS Therm Mgn processor (3.2) • disk_read_disk_time_sda • • intel_rdt_value_ipc_nan_50 • • • irq_value_TLB • intel_pmu_value_branches_42 • intel_rdt_value_ipc_nan_53 • • • load_longterm • intel_pmu_value_instructions_14 • intel_rdt_value_ipc_nan_54 • • • load_midterm • intel_pmu_value_page-faults_51 • intel_rdt_value_ipc_nan_9 • load_shortterm • intel_pmu_value_page-faults_all • numa_value_other_node_node1 • • • memory_value_cached • • memory_value_slab_unrecl Top TopTop TopTelemetryFeaturesSelectedByML 14 Recursive feature elimination stepping down the target from 40  20  10
  • 15. Featureselectionresults 15 • Applying various feature selection algorithms to reduce the number of telemetry data sampled without compromising the prediction accuracy • Smaller set of telemetry data saves training/inference time • vEPC test data set with original feature dimension ~400 • GradientBoostRegressor algorithm used for packet loss rate prediction • Select top number of features from feature importance output of GradientBoostRegressor
  • 16. KPIprediction&forecasting–supervisedlearning 16 • Regression • Train the model to predict target KPI using reduced telemetry data samples • Example: Packet drop rate • Accuracy Measured by MSE (Mean Squared Error) – the smaller the better • Classification • Detect packet drop from telemetry data • Measurement • Accuracy_score = correct_prediction / total_sample • Precision/Recall/F1-score, etc. • Time Series Forecasting • Predict future KPI value based on historical telemetry data and observed KPI trend
  • 17. KPIforecastingusingLSTM 17 • vCMTS downlink test data • KPI – scheduling packet loss rate • Input: previous 60 seconds of 20 selected telemetry data + KPI • Output: predicted future value (5 seconds later) of target KPI • Tensorflow BasicLSTMCell: Two layers, each layer 150 neurons
  • 18. Closed-loopAutomation Xeon vCMTS - 0 Traffic Generator vCMTS - 0VNFs NIC Machine Learning Modules collectd InfluxDB Traffic forecasting Reinforcement learning HW Optimization (RDT, DVFS, WL consolidation, etc.) Compacted metrics Using ML: • Track/forecast workload, performance • Dynamically adjust resource allocation Benefits: • Reduced TCO through power saving and increased HW utilization
  • 20. ENIPoC–NetworkSliceLifecycleManagement 20 • For generating new scale up/down and converting the intent to suggested configuration • LSTM is used for traffic prediction AI-based predictor: TNSM: CNSM: • Provides underlay network control to satisfy the network slice requests • FlexE and a FlexE-based optimization algorithm are used for underlay network slice creation and modification • Provides core network control to satisfy the network slice requests ETSI ISG ENI – Experiential Networked Intelligence
  • 24. Further Resources Learn more from these helpful sites: https://networkbuilders.intel.com/network-technologies/serviceassurance https://wiki.opnfv.org/display/fastpath/Barometer+Home https://wiki.openstack.org/wiki/Telemetry https://01.org/openstack/blogs/2015/openstack-enhanced-platform- awareness-white-paper
  • 25. 25 Collectd101materials • Collectd 101 • https://wiki.opnfv.org/display/fastpath/Collectd+101 • Write simple read plugin • https://wiki.opnfv.org/display/fastpath/Collectd+how+to+implement+a+si mple+plugin
  • 26. 26 Barometer Strategy: • Ensure platform metrics/events are accessible through open industry standard interfaces. • Demonstrate IA platform technologies can be monitored, consumed and actioned in real time Opnfvbarometer–Intelplatformfeatureplugins One Click Install: • Easy install/configuration for customers • One command to install Collectd/Influxdb/Grafana Three container approach for Collectd: • Stable Container: latest stable branch • Master Container: up to date with master • Experimental Container: cherry pick features of interest
  • 27. BarometerLinks Barometer Home: https://wiki.opnfv.org/display/fastpath/Barometer+Home Metrics/Events through Barometer (not on Collectd site): https://wiki.opnfv.org/display/fastpath/Collectd+Metrics+and+Events#CollectdM etricsandEvents-Metrics Barometer “One-click” install: https://wiki.opnfv.org/display/fastpath/One+Click+Install+of+Barometer+Contain ers 27