SlideShare a Scribd company logo
Practical ,Transparent Operating System Support For

 SUPERPAGES

Authors:
Juan Navarro, Sitaram Iyer, Peter Drusche, Alan Cox

Presented by:
Nadeeshani Hewage
Basics & Terminology
•   TLB?
     o Caches virtual-to-physical address mappings from the page tables
     o Faster access
•   TLB Coverage?
     o Amount of memory accessible through cached mappings, without TLB
       misses
•   Superpage
     o Memory page for larger size than a Base-page (ordinary memory page)
     o Multiple sizes
     o Single entry in TLB per superpage
•   Memory objects
     o Occupy some contiguous space in virtual memory
     o Eg: memory mapped files, code, data
•   TLB coverage decreasing with growing memory size over years
•   Motivation to increase TLB coverage without enlargening TLB size.
Constraints for Superpages
•   Page size should be supported by processor.

•   Contiguous

•   Starting address in physical & virtual address space must be a multiple of its
    size.

•   Single TLB entry for a superpage. Meaning single set of protection
    attributes and single dirty bit for a superpage.
Considerations
•   Allocation
     o Allocating physical frames for the memory object.
     o If there exists a plan to create a superpage by the OS, to satisfy the
        contiguity requirement,
           Relocation based allocation
              •  Create contiguity
              •  Physically copying allocated page frames when it is decided
                 that superpages are likely to be beneficial
                    o Software managed TLB + TLB miss handler
                    o Counter per superpage incremented at TLB miss
           Reservation based allocation
              •  Preserve contiguity
              •  Make allocation decisions at page fault time
              •  Decide preferred size of allocation on an algorithm
                    o Eg: Talluri & Hill proposed
•   Fragmentation Control
     o Release contiguous chunks of inactive memory from previous
       allocations
     o Preempt an existing partially used reservation

•   Promotion
     o Set of base pages from a potential super page satisfying constraints
       promoted to a Superpage.

•   Demotion
     o Reduce size of a superpage
           A smaller super page
           Base pages
•   Eviction
     o At a demand of memory preassure, inactive superpage is evicted from
        physical memory.
Design & Implementation
Preferred Superpage Size Policy
  o Decision made early at process execution
  o Looks only at attributes of memory object
  o Decided size
      Too large: decision overridden by preempting init reservation
      Too small: cannot be reverted
  o Policy: Pick the maximum superpage size that can be used for an object
Reservation based Allocation
•   Already spoken
Preempting Reservations
•   Free physical memory scarce/ excessively fragmented

•   Preempt reserved but unused frames

•   Choice between
     o Refusing allocation; reserving smaller extent than desired for new
       allocation
     o Preempting existing reservation which has enough unallocated frames

•   Policy: Whenever possible preempt & give space for alloaction
Incremental Promotions, Demotions
•   Promotions
     o Multiple superpage sizes exist
     o Superpage gets created when any size supported gets fully populated
       within a reservation
     o Increment smaller superpage to next larger size superpage

•   Demotions
     o A base page of a superpage targetted to be evicted by page daemon,
       causes a demotion

     o Demote to next smaller superpage size
Paging Out Dirty Superpages
•   OS keeps only single dirty bit for the whole superpage.

•   Problem: only one base page modified; cost of writing full superpage to disk
    (high cost on I/O)

•   Practice: Demote clean superpages whenever process attempts to write to
    them. Repromote later.
Reservation Lists
•   Keep track of reserved page extent that are not fully populated

•   One list per each page size supported.

•   Kept sorted by the time of their most recent page frame allocation

•   Policy: Preempt the extent whose most recent allocation, occurred least
    recently among all reservations in the list (head)
Population Map
•   Keep track of allocated base pages within each memory object
•   Queried on every page fault, should support efficient lookups.
•   Radix tree
     o Each level corresponds to a page size
     o Root: max superpage size supported
     o Non-leaf:
          # of supperpage sized virtual regions at next lower level with a
            population of at least one (somepop)
          # of supperpage sized virtual regions at next lower level fully
            populated (fullpop)
•   Used for various decisions
Wired Page clustering
•   Pages used for internal datastructures for the OS (implementation on
    FreeBSD) are wired
    (not evicted)

•   At system boot time, clustered together in physical memory.

•   Gradually as kernel allocates memory they get scattered, causes
    fragmentation

•   To avoid fragmentation, identify pages which are to be wired (internal use),
    cluster them in pools of contiguous physical memory
Evaluation
Platform & Workload
•   Platform
     o FreeBSD 4.3 kernel as a loadable module
     o Alpha 21264 processor at 500 MHz
     o four page sizes: 8KB base pages, 64KB, 512KB and 4MB superpages
     o 512MB RAM
     o 64KB data and 64KB instruction L1 caches, virtually indexed and 2-way
        associative
     o fully associative TLB with 128 entries for data and 128 for instructions
•   Workload
     o FFTW: The Fastest Fourier Transform in the West with a 200x200x200
        matrix as input
     o Matrix: A non-blocked matrix transposition of a1000x1000 matrix
     o Web: The thttpd web server servicing 50000 requests selected from an
        access log of the CS departmental web server at Rice University. The
        working set size of this trace is 238MB, while its data set is 3.6GB.
Best case benefits
•   When free memory is plentiful
    & non fragmented
     o Speedup
       (mean elapsed runtime of 3
       runs after an initial warmup)
     o TLB misses reduction
       percentage
Evaluation ctd.
•   Fragment system by running web server & feeding it with requests
•   FFTW run on fragmented system 4 times in sequence.
•   Goal : To see how quickly the system recovers just enough contiguous
    memory to build superpages
•   2 contiguity restoration techniques
     o Cache scheme
          Treat all cache pages as available, coalesces them into allocator
          0 superpages
     o Daemon scheme
          Contiguity aware page replacement + wired page clustering
          Get 20/60, 35/60, 38/60, 40/60 superpages
•   Even if the superpages implementation is there
    it requires a promising fragmentation control
    mechanism to acquire promising results.
•   With superpages and effective fragmentation control, evaluations have
    given promising results.
Thank You!

More Related Content

What's hot

google file system
google file systemgoogle file system
google file systemdiptipan
 
Operating System-Memory Management
Operating System-Memory ManagementOperating System-Memory Management
Operating System-Memory ManagementAkmal Cikmat
 
chapter 1 intoduction to operating system
chapter 1 intoduction to operating systemchapter 1 intoduction to operating system
chapter 1 intoduction to operating systemSiddhi Viradiya
 
Introduction to Operating System (Important Notes)
Introduction to Operating System (Important Notes)Introduction to Operating System (Important Notes)
Introduction to Operating System (Important Notes)Gaurav Kakade
 
Window scheduling algorithm
Window scheduling algorithmWindow scheduling algorithm
Window scheduling algorithmBinal Parekh
 
Translation Look Aside buffer
Translation Look Aside buffer Translation Look Aside buffer
Translation Look Aside buffer Zara Nawaz
 
Paging and Segmentation in Operating System
Paging and Segmentation in Operating SystemPaging and Segmentation in Operating System
Paging and Segmentation in Operating SystemRaj Mohan
 
8 memory management strategies
8 memory management strategies8 memory management strategies
8 memory management strategiesDr. Loganathan R
 
Real time operating system
Real time operating systemReal time operating system
Real time operating systemPratik Hiremath
 
Kernel. Operating System
Kernel. Operating SystemKernel. Operating System
Kernel. Operating Systempratikkadam78
 
Multi-Threading.pptx
Multi-Threading.pptxMulti-Threading.pptx
Multi-Threading.pptxCHANDRUG31
 

What's hot (20)

google file system
google file systemgoogle file system
google file system
 
Operating System-Memory Management
Operating System-Memory ManagementOperating System-Memory Management
Operating System-Memory Management
 
chapter 1 intoduction to operating system
chapter 1 intoduction to operating systemchapter 1 intoduction to operating system
chapter 1 intoduction to operating system
 
Introduction to Operating System (Important Notes)
Introduction to Operating System (Important Notes)Introduction to Operating System (Important Notes)
Introduction to Operating System (Important Notes)
 
Window scheduling algorithm
Window scheduling algorithmWindow scheduling algorithm
Window scheduling algorithm
 
Virtual memory
Virtual memoryVirtual memory
Virtual memory
 
Linux Memory Management
Linux Memory ManagementLinux Memory Management
Linux Memory Management
 
Memory Management
Memory ManagementMemory Management
Memory Management
 
Translation Look Aside buffer
Translation Look Aside buffer Translation Look Aside buffer
Translation Look Aside buffer
 
Fragmentaton
Fragmentaton Fragmentaton
Fragmentaton
 
Paging and Segmentation in Operating System
Paging and Segmentation in Operating SystemPaging and Segmentation in Operating System
Paging and Segmentation in Operating System
 
Storage Management
Storage ManagementStorage Management
Storage Management
 
8 memory management strategies
8 memory management strategies8 memory management strategies
8 memory management strategies
 
Bus Based Multiprocessors v2
Bus Based Multiprocessors v2Bus Based Multiprocessors v2
Bus Based Multiprocessors v2
 
VIRTUAL MEMORY
VIRTUAL MEMORYVIRTUAL MEMORY
VIRTUAL MEMORY
 
Mass storage systemsos
Mass storage systemsosMass storage systemsos
Mass storage systemsos
 
Ch5 answers
Ch5 answersCh5 answers
Ch5 answers
 
Real time operating system
Real time operating systemReal time operating system
Real time operating system
 
Kernel. Operating System
Kernel. Operating SystemKernel. Operating System
Kernel. Operating System
 
Multi-Threading.pptx
Multi-Threading.pptxMulti-Threading.pptx
Multi-Threading.pptx
 

Similar to Practical ,Transparent Operating System Support For Superpages

Java one2015 - Work With Hundreds of Hot Terabytes in JVMs
Java one2015 - Work With Hundreds of Hot Terabytes in JVMsJava one2015 - Work With Hundreds of Hot Terabytes in JVMs
Java one2015 - Work With Hundreds of Hot Terabytes in JVMsSpeedment, Inc.
 
HBase Sizing Guide
HBase Sizing GuideHBase Sizing Guide
HBase Sizing Guidelarsgeorge
 
The life and times
The life and timesThe life and times
The life and timesAbeer Naskar
 
Understanding memory management
Understanding memory managementUnderstanding memory management
Understanding memory managementGokul Vasan
 
Operating systems- Main Memory Management
Operating systems- Main Memory ManagementOperating systems- Main Memory Management
Operating systems- Main Memory ManagementChandrakant Divate
 
Dissecting Scalable Database Architectures
Dissecting Scalable Database ArchitecturesDissecting Scalable Database Architectures
Dissecting Scalable Database Architectureshypertable
 
Unit-4 swapping.pptx
Unit-4 swapping.pptxUnit-4 swapping.pptx
Unit-4 swapping.pptxItechAnand1
 
In-memory Data Management Trends & Techniques
In-memory Data Management Trends & TechniquesIn-memory Data Management Trends & Techniques
In-memory Data Management Trends & TechniquesHazelcast
 
Driver development – memory management
Driver development – memory managementDriver development – memory management
Driver development – memory managementVandana Salve
 
Training Webinar: Enterprise application performance with distributed caching
Training Webinar: Enterprise application performance with distributed cachingTraining Webinar: Enterprise application performance with distributed caching
Training Webinar: Enterprise application performance with distributed cachingOutSystems
 
Lecture 8- Virtual Memory Final.pptx
Lecture 8- Virtual Memory Final.pptxLecture 8- Virtual Memory Final.pptx
Lecture 8- Virtual Memory Final.pptxAmanuelmergia
 

Similar to Practical ,Transparent Operating System Support For Superpages (20)

Java one2015 - Work With Hundreds of Hot Terabytes in JVMs
Java one2015 - Work With Hundreds of Hot Terabytes in JVMsJava one2015 - Work With Hundreds of Hot Terabytes in JVMs
Java one2015 - Work With Hundreds of Hot Terabytes in JVMs
 
virtual_memory (3).pptx
virtual_memory (3).pptxvirtual_memory (3).pptx
virtual_memory (3).pptx
 
virtual memory
virtual memoryvirtual memory
virtual memory
 
Chapter08
Chapter08Chapter08
Chapter08
 
HBase Sizing Guide
HBase Sizing GuideHBase Sizing Guide
HBase Sizing Guide
 
Virtual Memory.pdf
Virtual Memory.pdfVirtual Memory.pdf
Virtual Memory.pdf
 
Os unit 3
Os unit 3Os unit 3
Os unit 3
 
The life and times
The life and timesThe life and times
The life and times
 
Understanding memory management
Understanding memory managementUnderstanding memory management
Understanding memory management
 
Operating systems- Main Memory Management
Operating systems- Main Memory ManagementOperating systems- Main Memory Management
Operating systems- Main Memory Management
 
Dissecting Scalable Database Architectures
Dissecting Scalable Database ArchitecturesDissecting Scalable Database Architectures
Dissecting Scalable Database Architectures
 
Segmentation and paging
Segmentation and paging Segmentation and paging
Segmentation and paging
 
Unit-4 swapping.pptx
Unit-4 swapping.pptxUnit-4 swapping.pptx
Unit-4 swapping.pptx
 
Memory Management.pdf
Memory Management.pdfMemory Management.pdf
Memory Management.pdf
 
Demand paging
Demand pagingDemand paging
Demand paging
 
In-memory Data Management Trends & Techniques
In-memory Data Management Trends & TechniquesIn-memory Data Management Trends & Techniques
In-memory Data Management Trends & Techniques
 
Driver development – memory management
Driver development – memory managementDriver development – memory management
Driver development – memory management
 
Training Webinar: Enterprise application performance with distributed caching
Training Webinar: Enterprise application performance with distributed cachingTraining Webinar: Enterprise application performance with distributed caching
Training Webinar: Enterprise application performance with distributed caching
 
UNIT IV.pptx
UNIT IV.pptxUNIT IV.pptx
UNIT IV.pptx
 
Lecture 8- Virtual Memory Final.pptx
Lecture 8- Virtual Memory Final.pptxLecture 8- Virtual Memory Final.pptx
Lecture 8- Virtual Memory Final.pptx
 

Recently uploaded

Neuro-symbolic is not enough, we need neuro-*semantic*
Neuro-symbolic is not enough, we need neuro-*semantic*Neuro-symbolic is not enough, we need neuro-*semantic*
Neuro-symbolic is not enough, we need neuro-*semantic*Frank van Harmelen
 
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualitySoftware Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualityInflectra
 
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Jeffrey Haguewood
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf91mobiles
 
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...Thierry Lestable
 
Connector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a buttonConnector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a buttonDianaGray10
 
When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...Elena Simperl
 
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...Product School
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesLaura Byrne
 
How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...Product School
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance
 
Assuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesAssuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesThousandEyes
 
UiPath New York Community Day in-person event
UiPath New York Community Day in-person eventUiPath New York Community Day in-person event
UiPath New York Community Day in-person eventDianaGray10
 
НАДІЯ ФЕДЮШКО БАЦ «Професійне зростання QA спеціаліста»
НАДІЯ ФЕДЮШКО БАЦ  «Професійне зростання QA спеціаліста»НАДІЯ ФЕДЮШКО БАЦ  «Професійне зростання QA спеціаліста»
НАДІЯ ФЕДЮШКО БАЦ «Професійне зростання QA спеціаліста»QADay
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Product School
 
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Ramesh Iyer
 
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Product School
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform EngineeringJemma Hussein Allen
 
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Product School
 

Recently uploaded (20)

Neuro-symbolic is not enough, we need neuro-*semantic*
Neuro-symbolic is not enough, we need neuro-*semantic*Neuro-symbolic is not enough, we need neuro-*semantic*
Neuro-symbolic is not enough, we need neuro-*semantic*
 
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualitySoftware Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
 
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
 
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
 
Connector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a buttonConnector Corner: Automate dynamic content and events by pushing a button
Connector Corner: Automate dynamic content and events by pushing a button
 
When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...
 
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
 
The Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and SalesThe Art of the Pitch: WordPress Relationships and Sales
The Art of the Pitch: WordPress Relationships and Sales
 
How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...
 
Ransomware Mallox [EN].pdf
Ransomware         Mallox       [EN].pdfRansomware         Mallox       [EN].pdf
Ransomware Mallox [EN].pdf
 
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdfFIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
FIDO Alliance Osaka Seminar: Passkeys at Amazon.pdf
 
Assuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesAssuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyes
 
UiPath New York Community Day in-person event
UiPath New York Community Day in-person eventUiPath New York Community Day in-person event
UiPath New York Community Day in-person event
 
НАДІЯ ФЕДЮШКО БАЦ «Професійне зростання QA спеціаліста»
НАДІЯ ФЕДЮШКО БАЦ  «Професійне зростання QA спеціаліста»НАДІЯ ФЕДЮШКО БАЦ  «Професійне зростання QA спеціаліста»
НАДІЯ ФЕДЮШКО БАЦ «Професійне зростання QA спеціаліста»
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...
 
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
Builder.ai Founder Sachin Dev Duggal's Strategic Approach to Create an Innova...
 
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
 
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
 

Practical ,Transparent Operating System Support For Superpages

  • 1. Practical ,Transparent Operating System Support For SUPERPAGES Authors: Juan Navarro, Sitaram Iyer, Peter Drusche, Alan Cox Presented by: Nadeeshani Hewage
  • 2. Basics & Terminology • TLB? o Caches virtual-to-physical address mappings from the page tables o Faster access • TLB Coverage? o Amount of memory accessible through cached mappings, without TLB misses • Superpage o Memory page for larger size than a Base-page (ordinary memory page) o Multiple sizes o Single entry in TLB per superpage • Memory objects o Occupy some contiguous space in virtual memory o Eg: memory mapped files, code, data • TLB coverage decreasing with growing memory size over years • Motivation to increase TLB coverage without enlargening TLB size.
  • 3. Constraints for Superpages • Page size should be supported by processor. • Contiguous • Starting address in physical & virtual address space must be a multiple of its size. • Single TLB entry for a superpage. Meaning single set of protection attributes and single dirty bit for a superpage.
  • 4. Considerations • Allocation o Allocating physical frames for the memory object. o If there exists a plan to create a superpage by the OS, to satisfy the contiguity requirement,  Relocation based allocation • Create contiguity • Physically copying allocated page frames when it is decided that superpages are likely to be beneficial o Software managed TLB + TLB miss handler o Counter per superpage incremented at TLB miss  Reservation based allocation • Preserve contiguity • Make allocation decisions at page fault time • Decide preferred size of allocation on an algorithm o Eg: Talluri & Hill proposed
  • 5. Fragmentation Control o Release contiguous chunks of inactive memory from previous allocations o Preempt an existing partially used reservation • Promotion o Set of base pages from a potential super page satisfying constraints promoted to a Superpage. • Demotion o Reduce size of a superpage  A smaller super page  Base pages • Eviction o At a demand of memory preassure, inactive superpage is evicted from physical memory.
  • 7. Preferred Superpage Size Policy o Decision made early at process execution o Looks only at attributes of memory object o Decided size  Too large: decision overridden by preempting init reservation  Too small: cannot be reverted o Policy: Pick the maximum superpage size that can be used for an object
  • 9. Preempting Reservations • Free physical memory scarce/ excessively fragmented • Preempt reserved but unused frames • Choice between o Refusing allocation; reserving smaller extent than desired for new allocation o Preempting existing reservation which has enough unallocated frames • Policy: Whenever possible preempt & give space for alloaction
  • 10. Incremental Promotions, Demotions • Promotions o Multiple superpage sizes exist o Superpage gets created when any size supported gets fully populated within a reservation o Increment smaller superpage to next larger size superpage • Demotions o A base page of a superpage targetted to be evicted by page daemon, causes a demotion o Demote to next smaller superpage size
  • 11. Paging Out Dirty Superpages • OS keeps only single dirty bit for the whole superpage. • Problem: only one base page modified; cost of writing full superpage to disk (high cost on I/O) • Practice: Demote clean superpages whenever process attempts to write to them. Repromote later.
  • 12. Reservation Lists • Keep track of reserved page extent that are not fully populated • One list per each page size supported. • Kept sorted by the time of their most recent page frame allocation • Policy: Preempt the extent whose most recent allocation, occurred least recently among all reservations in the list (head)
  • 13. Population Map • Keep track of allocated base pages within each memory object • Queried on every page fault, should support efficient lookups. • Radix tree o Each level corresponds to a page size o Root: max superpage size supported o Non-leaf:  # of supperpage sized virtual regions at next lower level with a population of at least one (somepop)  # of supperpage sized virtual regions at next lower level fully populated (fullpop) • Used for various decisions
  • 14.
  • 15. Wired Page clustering • Pages used for internal datastructures for the OS (implementation on FreeBSD) are wired (not evicted) • At system boot time, clustered together in physical memory. • Gradually as kernel allocates memory they get scattered, causes fragmentation • To avoid fragmentation, identify pages which are to be wired (internal use), cluster them in pools of contiguous physical memory
  • 17. Platform & Workload • Platform o FreeBSD 4.3 kernel as a loadable module o Alpha 21264 processor at 500 MHz o four page sizes: 8KB base pages, 64KB, 512KB and 4MB superpages o 512MB RAM o 64KB data and 64KB instruction L1 caches, virtually indexed and 2-way associative o fully associative TLB with 128 entries for data and 128 for instructions • Workload o FFTW: The Fastest Fourier Transform in the West with a 200x200x200 matrix as input o Matrix: A non-blocked matrix transposition of a1000x1000 matrix o Web: The thttpd web server servicing 50000 requests selected from an access log of the CS departmental web server at Rice University. The working set size of this trace is 238MB, while its data set is 3.6GB.
  • 18. Best case benefits • When free memory is plentiful & non fragmented o Speedup (mean elapsed runtime of 3 runs after an initial warmup) o TLB misses reduction percentage
  • 19. Evaluation ctd. • Fragment system by running web server & feeding it with requests • FFTW run on fragmented system 4 times in sequence. • Goal : To see how quickly the system recovers just enough contiguous memory to build superpages • 2 contiguity restoration techniques o Cache scheme  Treat all cache pages as available, coalesces them into allocator  0 superpages o Daemon scheme  Contiguity aware page replacement + wired page clustering  Get 20/60, 35/60, 38/60, 40/60 superpages
  • 20. Even if the superpages implementation is there it requires a promising fragmentation control mechanism to acquire promising results.
  • 21. With superpages and effective fragmentation control, evaluations have given promising results.