Submit Search
Upload
Generic Resource Manager - László Vadkerti, András Kovács
•
1 like
•
908 views
H
harryvanhaaren
Follow
Generic Resource Manager
Read less
Read more
Technology
Report
Share
Report
Share
1 of 36
Download now
Download to read offline
Recommended
DPDK Integration: A Product's Journey - Roger B. Melton
DPDK Integration: A Product's Journey - Roger B. Melton
harryvanhaaren
DPDK Summit - 08 Sept 2014 - Ericsson - A Multi-Socket Ferrari for NFV
DPDK Summit - 08 Sept 2014 - Ericsson - A Multi-Socket Ferrari for NFV
Jim St. Leger
Hyperscan - Mohammad Abdul Awal
Hyperscan - Mohammad Abdul Awal
harryvanhaaren
DPDK Architecture Musings - Andy Harvey
DPDK Architecture Musings - Andy Harvey
harryvanhaaren
Hotplug and Virtio - Tetsuya Mukawa
Hotplug and Virtio - Tetsuya Mukawa
harryvanhaaren
OpenDataPlane - Bill Fischofer
OpenDataPlane - Bill Fischofer
harryvanhaaren
Cloud Networking Trends
Cloud Networking Trends
Michelle Holley
Symmetric Crypto for DPDK - Declan Doherty
Symmetric Crypto for DPDK - Declan Doherty
harryvanhaaren
Recommended
DPDK Integration: A Product's Journey - Roger B. Melton
DPDK Integration: A Product's Journey - Roger B. Melton
harryvanhaaren
DPDK Summit - 08 Sept 2014 - Ericsson - A Multi-Socket Ferrari for NFV
DPDK Summit - 08 Sept 2014 - Ericsson - A Multi-Socket Ferrari for NFV
Jim St. Leger
Hyperscan - Mohammad Abdul Awal
Hyperscan - Mohammad Abdul Awal
harryvanhaaren
DPDK Architecture Musings - Andy Harvey
DPDK Architecture Musings - Andy Harvey
harryvanhaaren
Hotplug and Virtio - Tetsuya Mukawa
Hotplug and Virtio - Tetsuya Mukawa
harryvanhaaren
OpenDataPlane - Bill Fischofer
OpenDataPlane - Bill Fischofer
harryvanhaaren
Cloud Networking Trends
Cloud Networking Trends
Michelle Holley
Symmetric Crypto for DPDK - Declan Doherty
Symmetric Crypto for DPDK - Declan Doherty
harryvanhaaren
FD.io - The Universal Dataplane
FD.io - The Universal Dataplane
Open Networking Summit
DPDK Summit 2015 - Sprint - Arun Rajagopal
DPDK Summit 2015 - Sprint - Arun Rajagopal
Jim St. Leger
LF_DPDK17_Implementation and Testing of Soft Patch Panel
LF_DPDK17_Implementation and Testing of Soft Patch Panel
LF_DPDK
Inside Microsoft's FPGA-Based Configurable Cloud
Inside Microsoft's FPGA-Based Configurable Cloud
inside-BigData.com
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
OPNFV
ODSA Use Case - SmartNIC
ODSA Use Case - SmartNIC
ODSA Workgroup
Dpdk Validation - Liu, Yong
Dpdk Validation - Liu, Yong
harryvanhaaren
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
Michelle Holley
TLDK - FD.io Sept 2016
TLDK - FD.io Sept 2016
Benoit Hudzia
LF_DPDK17_mediated devices: better userland IO
LF_DPDK17_mediated devices: better userland IO
LF_DPDK
Netsft2017 day in_life_of_nfv
Netsft2017 day in_life_of_nfv
Intel
DPDK Summit 2015 - Aspera - Charles Shiflett
DPDK Summit 2015 - Aspera - Charles Shiflett
Jim St. Leger
Advanced Traffic Engineering (TE++)
Advanced Traffic Engineering (TE++)
Pravin Bhandarkar
Building a Router
Building a Router
Hannes Gredler
Tungsten Fabric Overview
Tungsten Fabric Overview
Michelle Holley
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
harryvanhaaren
The Need for Complex Analytics from Forwarding Pipelines
The Need for Complex Analytics from Forwarding Pipelines
Netronome
Performance challenges in software networking
Performance challenges in software networking
Stephen Hemminger
DPDK IPSec performance benchmark ~ Georgii Tkachuk
DPDK IPSec performance benchmark ~ Georgii Tkachuk
Intel
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
LF_DPDK
Emc vspex customer_presentation_private_cloud_virtualized_share_point
Emc vspex customer_presentation_private_cloud_virtualized_share_point
xKinAnx
Elastify Cloud-Native Spark Application with Persistent Memory
Elastify Cloud-Native Spark Application with Persistent Memory
Databricks
More Related Content
What's hot
FD.io - The Universal Dataplane
FD.io - The Universal Dataplane
Open Networking Summit
DPDK Summit 2015 - Sprint - Arun Rajagopal
DPDK Summit 2015 - Sprint - Arun Rajagopal
Jim St. Leger
LF_DPDK17_Implementation and Testing of Soft Patch Panel
LF_DPDK17_Implementation and Testing of Soft Patch Panel
LF_DPDK
Inside Microsoft's FPGA-Based Configurable Cloud
Inside Microsoft's FPGA-Based Configurable Cloud
inside-BigData.com
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
OPNFV
ODSA Use Case - SmartNIC
ODSA Use Case - SmartNIC
ODSA Workgroup
Dpdk Validation - Liu, Yong
Dpdk Validation - Liu, Yong
harryvanhaaren
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
Michelle Holley
TLDK - FD.io Sept 2016
TLDK - FD.io Sept 2016
Benoit Hudzia
LF_DPDK17_mediated devices: better userland IO
LF_DPDK17_mediated devices: better userland IO
LF_DPDK
Netsft2017 day in_life_of_nfv
Netsft2017 day in_life_of_nfv
Intel
DPDK Summit 2015 - Aspera - Charles Shiflett
DPDK Summit 2015 - Aspera - Charles Shiflett
Jim St. Leger
Advanced Traffic Engineering (TE++)
Advanced Traffic Engineering (TE++)
Pravin Bhandarkar
Building a Router
Building a Router
Hannes Gredler
Tungsten Fabric Overview
Tungsten Fabric Overview
Michelle Holley
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
harryvanhaaren
The Need for Complex Analytics from Forwarding Pipelines
The Need for Complex Analytics from Forwarding Pipelines
Netronome
Performance challenges in software networking
Performance challenges in software networking
Stephen Hemminger
DPDK IPSec performance benchmark ~ Georgii Tkachuk
DPDK IPSec performance benchmark ~ Georgii Tkachuk
Intel
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
LF_DPDK
What's hot
(20)
FD.io - The Universal Dataplane
FD.io - The Universal Dataplane
DPDK Summit 2015 - Sprint - Arun Rajagopal
DPDK Summit 2015 - Sprint - Arun Rajagopal
LF_DPDK17_Implementation and Testing of Soft Patch Panel
LF_DPDK17_Implementation and Testing of Soft Patch Panel
Inside Microsoft's FPGA-Based Configurable Cloud
Inside Microsoft's FPGA-Based Configurable Cloud
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
Fast datastacks - fast and flexible nfv solution stacks leveraging fd.io
ODSA Use Case - SmartNIC
ODSA Use Case - SmartNIC
Dpdk Validation - Liu, Yong
Dpdk Validation - Liu, Yong
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
Install FD.IO VPP On Intel(r) Architecture & Test with Trex*
TLDK - FD.io Sept 2016
TLDK - FD.io Sept 2016
LF_DPDK17_mediated devices: better userland IO
LF_DPDK17_mediated devices: better userland IO
Netsft2017 day in_life_of_nfv
Netsft2017 day in_life_of_nfv
DPDK Summit 2015 - Aspera - Charles Shiflett
DPDK Summit 2015 - Aspera - Charles Shiflett
Advanced Traffic Engineering (TE++)
Advanced Traffic Engineering (TE++)
Building a Router
Building a Router
Tungsten Fabric Overview
Tungsten Fabric Overview
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
OVS and DPDK - T.F. Herbert, K. Traynor, M. Gray
The Need for Complex Analytics from Forwarding Pipelines
The Need for Complex Analytics from Forwarding Pipelines
Performance challenges in software networking
Performance challenges in software networking
DPDK IPSec performance benchmark ~ Georgii Tkachuk
DPDK IPSec performance benchmark ~ Georgii Tkachuk
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
LF_DPDK17_Accelerating NFV with VMware's Enhanced Network Stack (ENS) and Int...
Similar to Generic Resource Manager - László Vadkerti, András Kovács
Emc vspex customer_presentation_private_cloud_virtualized_share_point
Emc vspex customer_presentation_private_cloud_virtualized_share_point
xKinAnx
Elastify Cloud-Native Spark Application with Persistent Memory
Elastify Cloud-Native Spark Application with Persistent Memory
Databricks
Volatile Uses for Persistent Memory
Volatile Uses for Persistent Memory
Intel® Software
Srm suite technical presentation nrm - tim piqueur
Srm suite technical presentation nrm - tim piqueur
EMC Nederland
Emc data domain technical deep dive workshop
Emc data domain technical deep dive workshop
solarisyougood
Optimize DR and Cloning with Logical Hostnames in Oracle E-Business Suite (OA...
Optimize DR and Cloning with Logical Hostnames in Oracle E-Business Suite (OA...
Andrejs Prokopjevs
NetBackup Appliance Family presentation
NetBackup Appliance Family presentation
Symantec
MIG 5th Data Centre Summit 2016 PTS Presentation v1
MIG 5th Data Centre Summit 2016 PTS Presentation v1
blewington
TECHNICAL BRIEF▶ Backup Exec 15 Blueprint for Large Installations
TECHNICAL BRIEF▶ Backup Exec 15 Blueprint for Large Installations
Symantec
BMC: Bare Metal Container @Open Source Summit Japan 2017
BMC: Bare Metal Container @Open Source Summit Japan 2017
Kuniyasu Suzaki
Emc sql server 2012 overview
Emc sql server 2012 overview
solarisyougood
Transforming Mission Critical Applications
Transforming Mission Critical Applications
Cenk Ersoy
Pro sphere customer technical
Pro sphere customer technical
solarisyougood
EMC Vmax3 tech-deck deep dive
EMC Vmax3 tech-deck deep dive
solarisyougood
Memory-Driven Near-Data Acceleration and its application to DOME/SKA
Memory-Driven Near-Data Acceleration and its application to DOME/SKA
inside-BigData.com
Accelerating SDN/NFV with transparent offloading architecture
Accelerating SDN/NFV with transparent offloading architecture
Open Networking Summits
NGD Systems and Microsoft Keynote Presentation at IPDPS MPP in Vacouver
NGD Systems and Microsoft Keynote Presentation at IPDPS MPP in Vacouver
Scott Shadley, MBA,PMC-III
Diablo Memory Channel Flash Podcast
Diablo Memory Channel Flash Podcast
inside-BigData.com
CtrlS: Cloud Solutions for Retail & eCommerce
CtrlS: Cloud Solutions for Retail & eCommerce
eTailing India
”Bare-Metal Container" presented at HPCC2016
”Bare-Metal Container" presented at HPCC2016
Kuniyasu Suzaki
Similar to Generic Resource Manager - László Vadkerti, András Kovács
(20)
Emc vspex customer_presentation_private_cloud_virtualized_share_point
Emc vspex customer_presentation_private_cloud_virtualized_share_point
Elastify Cloud-Native Spark Application with Persistent Memory
Elastify Cloud-Native Spark Application with Persistent Memory
Volatile Uses for Persistent Memory
Volatile Uses for Persistent Memory
Srm suite technical presentation nrm - tim piqueur
Srm suite technical presentation nrm - tim piqueur
Emc data domain technical deep dive workshop
Emc data domain technical deep dive workshop
Optimize DR and Cloning with Logical Hostnames in Oracle E-Business Suite (OA...
Optimize DR and Cloning with Logical Hostnames in Oracle E-Business Suite (OA...
NetBackup Appliance Family presentation
NetBackup Appliance Family presentation
MIG 5th Data Centre Summit 2016 PTS Presentation v1
MIG 5th Data Centre Summit 2016 PTS Presentation v1
TECHNICAL BRIEF▶ Backup Exec 15 Blueprint for Large Installations
TECHNICAL BRIEF▶ Backup Exec 15 Blueprint for Large Installations
BMC: Bare Metal Container @Open Source Summit Japan 2017
BMC: Bare Metal Container @Open Source Summit Japan 2017
Emc sql server 2012 overview
Emc sql server 2012 overview
Transforming Mission Critical Applications
Transforming Mission Critical Applications
Pro sphere customer technical
Pro sphere customer technical
EMC Vmax3 tech-deck deep dive
EMC Vmax3 tech-deck deep dive
Memory-Driven Near-Data Acceleration and its application to DOME/SKA
Memory-Driven Near-Data Acceleration and its application to DOME/SKA
Accelerating SDN/NFV with transparent offloading architecture
Accelerating SDN/NFV with transparent offloading architecture
NGD Systems and Microsoft Keynote Presentation at IPDPS MPP in Vacouver
NGD Systems and Microsoft Keynote Presentation at IPDPS MPP in Vacouver
Diablo Memory Channel Flash Podcast
Diablo Memory Channel Flash Podcast
CtrlS: Cloud Solutions for Retail & eCommerce
CtrlS: Cloud Solutions for Retail & eCommerce
”Bare-Metal Container" presented at HPCC2016
”Bare-Metal Container" presented at HPCC2016
Recently uploaded
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Edi Saputra
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
Zilliz
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptx
Remote DBA Services
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
Dropbox
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
Khushali Kathiriya
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
danishmna97
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Victor Rentea
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
Sandro Moreira
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
apidays
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
The Digital Insurer
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Orbitshub
Architecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
Zilliz
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
MIND CTI
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
apidays
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Orbitshub
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Zilliz
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
Remote DBA Services
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
Christopher Logan Kennedy
Recently uploaded
(20)
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
Vector Search -An Introduction in Oracle Database 23ai.pptx
Vector Search -An Introduction in Oracle Database 23ai.pptx
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Architecting Cloud Native Applications
Architecting Cloud Native Applications
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
Generic Resource Manager - László Vadkerti, András Kovács
1.
generic resource manager András KovÁcs (Andras.kovacs@ericsson.com) LászlÓ
Vadkerti (laszlo.vadkerti@Ericsson.com) A manager we would like :)
2.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 2 Last year…
3.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 3 by merriam-webster: › something that a country has and can use to increase its wealth › a supply of something (such as money) that someone has and can use when it is needed › a place or thing that provides something useful by wikipedia (computing): › A resource, or system resource, is any physical or virtual component of limited availability within a computer system. [...] Every internal system component is a resource. […]. What is a resource?
4.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 4 › CPU › Memory (hugepage, cache, ivshm, memzone) › Virtual address › Packet pool › Network/Virtual Device (port/queue) › Lcore › Instance › HW accelerator › And there are lots of other resources coming… (threads, protocol stacks, etc) typical dpdk resources
5.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 5 › Better control over resources › Hide “real” environment from applications › Support migration › Keep track of resources/usage (High Availability) › Prioritize resources › Access control › Avoid code duplication Why?
6.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 6 › Resource pools (resources with similar attributes) e.g. Servers -> Compute nodes -> Virtual Machines -> CPU/memory/interface/lcore/pktpool Resource pools MEM CPU I/F CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU
7.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 7 MEM CPU MEM CPU MEM CPU I/F › Resource pools (resources with similar attributes) e.g. Servers -> Compute nodes -> Virtual Machines -> CPU/memory/interface/lcore/pktpool Resource pools CPU CPU CPU CPU
8.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 8 MEM CPU MEM CPU MEM CPU I/F › Resource pools (resources with similar attributes) e.g. Servers -> Compute nodes -> Virtual Machines -> CPU/memory/interface/lcore/pktpool Resource pools CPU CPU attribute 1 attribute 2
9.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 9 › cpualias foreground { › cpumask = "2" › } › pktpool pktpool_fg { › cpualias = foreground › num_pkts = 8K › num_cached_pkts = 64 › } Config EXAMPLE PKT POOL CPU
10.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 10 › cpualias foreground { › cpumask = "2,12" › } › pktpool pktpool_fg { › cpualias = foreground › num_pkts = 8K › num_cached_pkts = 64 › } Config EXAMPLE PKT POOL CPU CPU S1S0
11.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 11 › cpualias foreground { › cpumask = "2,12" › } › pktpool pktpool_fg { › cpualias = foreground › num_pkts = 8K › num_cached_pkts = 64 type = numa › } Config EXAMPLE S1S0 PKT POOL PKT POOL CPU CPU
12.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 12 › cpualias foreground { › cpumask = "2,4,12,14" › } › pktpool pktpool_fg { › cpualias = foreground › num_pkts = 8K › num_cached_pkts = 64 type = numa › } Config EXAMPLE S1S0 PKT POOL PKT POOL CPU CPU CPUCPU
13.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 13 › cpualias foreground { › cpumask = "2,4,12,14" › } › pktpool pktpool_fg { › cpualias = foreground › num_pkts = 8K › num_cached_pkts = 64 type = exclusive › } Config EXAMPLE PKT POOL S1S0 PKT POOL PKT POOL PKT POOL CPU CPU CPUCPU
14.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 14 › resource allocation (static or on-demand) › chain of pools in case of dynamic allocation (example) RESOURCE allocation
15.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 16 › e.g. a native workstation with 4 sockets, memory, interfaces. › resmgr starts, autodetects (CPUs, HT siblings, cache sizes, etc) every resources we can use. › resmgr: daemon/binary/library › reading configuration and creating a free resource database (our choice is xattr) › supporting popup cpu, memory, interface, etc › linking/referencing resources (cpu table, HT table, NUMA, L3 cache (CAT/cache QoS), etc) RESMGR at work
16.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 17 › Is the relation between virtual ports, virtual queues and named packet pools. Somewhat similar to pipeline model configuration. packet domain
17.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 18 › Allocating resources › Per-application configuration › Application instances see only assigned resources (like a VM) DPDK DOMAIN (APPlication DOMAIN)
18.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 19 existing Memory management example process1 process2 1G 1G 1G 1G 1G 1G 1G 1G DPDK memory DPDK socket
19.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 20 › Simple memory management API › Transparent NUMA awareness for applications – Even in a non-NUMA aware guest VM environment › Flexible, configuration, re-configuration › Provide a more granular way of placing objects in memory – e.g. to control the number of TLB cache entry usage › Memory partitioning – Physically contiguous memory, hugepages etc. – Partitioning across application instances – Shared memory support memory management goals
20.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 21 › Named and indexed memory partition (shared memory) with defined properties such as – Location, e.g. NUMA node(s) – Size of partition, e.g. 4GB/1G, 512MB/2M hugepages – Physical contiguity – Type (future improvement), e.g. › Hugepage memory (only possible type today) › Mmap-ped file › Inter-vm shared memory › Distributed memory › etc. what is a memdomain?
21.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 22 › Is a pool of shared memory resources (memdomains) › Is a named group of memory partitions with different location attributes › Three main types – DEFAULT › No location preference, only a single memdomain in the pool – NUMA › One memdomain per involved NUMA node – EXCLUSIVE › One memdomain per application instance what is a memdomain pool?
22.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 23 MEMDOMAIN Configuration › Can be compared to partitioning a hard drive where – Available hugepage memory is the hard drive, e.g. one drive per NUMA node – Memdomains can be compared to disk partitions – Memzones can be compared to files › Should be a system engineering task to find the best model working for the specific application model
23.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 24 MEMDOMAIN Simple Application API › Initialize library (oai_memcfg_libinit) – Register a unique name to the process (High Availability, debug) – Initializes DPDK memory, grabs hugepages etc. › Attach to named memory partition (oai_memcfg_attach) – Using pre-configured memdomain pool name – Automatic memdomain selection from pool (per instance, NUMA) – Maps memdomain (memzones) into virtual address space
24.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 25 MEMDOMAIN ConfiguratioN EXAMPLE CPUALIAS cpualias all { cpumask = "0-31" # all available CPU } CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU
25.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 26 MEMDOMAIN ConfiguratioN EXAMPLE CPUALIAS cpualias all { cpumask = "0-31" # all available CPU } cpualias foreground { cpumask = "2-31:2" # even cpus excluding cpu0 } CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU CPU
26.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 27 MEMDOMAIN CONFIGURATION EXAMPLE DEFAULT TYPE # ------------------------------- # Shared memory accessed by # all instances # ------------------------------- memdomain App_Shared_Memory { type = default cpualias = "all" alloc_memzone = true size { huge_2M = 512M huge_1G = 0 } } Instance1 S1S0 MEM DOMAIN Instance2
27.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 28 memdomain API ATTACH EXAMPLE DEFAULT TYPE Instance1: - oai_memcfg_libinit(“AppProcess1”); - addr1 = oai_memcfg_attach(“App_Shared_Memory”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); Instance2: - oai_memcfg_libinit(“AppProcess2”); - addr2 = oai_memcfg_attach(“App_Shared_Memory”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); › Explanation: – addr1 in process1 will point to the same memory as addr2 in process2
28.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 29 MEMDOMAIN CONFIGURATION EXAMPLE NUMA TYPE # ---------------------------------------- # Per NUMA node shared memory. # App instances tied to the same NUMA # are sharing the same memory partition. # ---------------------------------------- memdomain App_NUMA_Shared { type = numa cpualias = "all" alloc_memzone = true size { is_per_numa = true huge_2M = 0 huge_1G = 1G } } Instance1 S1S0 MEM DOMAIN MEM DOMAIN Instance2 Instance3
29.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 30 memdomain API ATTACH EXAMPLE NUMA TYPE Instance1: - oai_memcfg_libinit(“AppProcess1”); - pin to vcpu on numa node0 (lcore_assign, pktpthread_setaffinity_np) - addr1 = oai_memcfg_attach(“App_NUMA_Shared”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); Instance2: - oai_memcfg_libinit(“AppProcess2”); - pin to vcpu on numa node1 (lcore_assign, pktpthread_setaffinity_np) - addr2 = oai_memcfg_attach(“App_NUMA_Shared”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); Instance3: - oai_memcfg_libinit(“AppProcess3”); - pin to vcpu on numa node1 (lcore_assign, pktpthread_setaffinity_np) - addr3 = oai_memcfg_attach(“App_NUMA_Shared”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); › Explanation: – addr1 in instance1 will point to the partition allocated in NUMA node0 while addr2 in instance2 and addr3 in instance3 will point to the same partition allocated in NUMA node1
30.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 31 memdomain CONFIGURATION EXAMPLE EXCLUSIVE TYPE # ----------------------------------- # Per App instance private memory. # Every instance has its own memory # partition. # ----------------------------------- memdomain App_Thread_Local { type = excl cpualias = "foreground" alloc_memzone = true size { is_per_cpu = true huge_2M = 0 huge_1G = 1G } } Instance1 S1S0 MEM DOMAIN MEM DOMAIN MEM DOMAIN Instance2 instance3
31.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 32 memdomain API ATTACH EXAMPLE EXCLUSIVE TYPE Instance1: - oai_memcfg_libinit(“AppProcess1”); - pin to vcpu2 on numa node0 (lcore_assign, pktpthread_setaffinity_np) - addr1 = oai_memcfg_attach(“App_Thread_Local”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); Instance2: - oai_memcfg_libinit(“AppProcess2”); - pin to vcpu16 on numa node1 (lcore_assign, pktpthread_setaffinity_np) - addr2 = oai_memcfg_attach(“App_Thread_Local”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); Instance3: - oai_memcfg_libinit(“AppProcess3”); - pin to vcpu18 on numa node1 (lcore_assign, pktpthread_setaffinity_np) - addr3 = oai_memcfg_attach(“App_Thread_Local”, OAI_MEMCFG_MD_INDEX_AUTO, &len, OAI_MEMCFG_MZ_FLAGS_NONE); › Explanation: – Addr1, addr2 and addr3 will point to different memory addresses – addr1 allocated on NUMA node0 while addr2 and addr3 allocated on NUMA node1
32.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 33 Advanced MEMCFG API › Create memdomain pools in a transaction – Creating a pool of memory partitions (partitions are auto aligned based on user's location or requested layout) – Creating multiple memdomain pools in a transaction allows better placement (e.g. prioritizing physically contiguous memory) › Attach to memdomain (memory partition) – Attach by memdomain name and index, returning md handle (index could be auto if instance location is known e.g. by lcoreid) – Could also MMAP whole partition into virtual address space › Get number of memory partitions in a memdomain pool – Allows attaching to all partitions by index within a pool e.g. for per NUMA table replication
33.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 34 Advanced MEMCFG API › Reserve named memzones by memdomain handle – Memzone name must only be unique within the memdomain, same name can be used in another memdomain e.g. it allows attaching to replicas using the same name – supports fragments if physical contiguity is not requested (virtually cont) › Lookup memzone by memdomain handle – Looks up memzone within a memdomain returning mz handle › Attach to memzone by memzone handle – Attaches to a memzone, e.g. mapping into requested/reserved virtual address space as readonly/readwrite
34.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 35 Memdomain example process1 process2 1G 1G 1G 1G 1G 1G 1G 1G DPDK memory memory partition 1 memory partition 2
35.
Generic resource manager
for DPDK | Public | © Ericsson AB 2015 | 2015-09-29 | Page 36 › Support non-DPDK applications › Memory types for rte_malloc › DPDK certification (for VM or hardware) › GUI or config file for dpdk startup parameters (RESMGR config alternative) › Visual resource allocator (GParted style) › DPDK OS / DPDK distro › standby instances › NEW emulated QEMU device Future ideas
Download now