Lesson 3b (Memory Virtualization)

Question 1

Why is virtual memory implementation so important

Answer 1

it is crucial to performance

Question 2

How many page tables do we have

Answer 2

one for each process

Question 3

Where is the pagetable (hardware or software)

Answer 3

it is an os data structure

Question 4

What does the page table contain

Answer 4

mappings from virtual to physical memory

Question 5

does the hypervisor know about the page table in each OS

Answer 5

no

Question 6

What is machine memory

Answer 6

physical memory, like the ram sticks in the machine

Question 7

who controls machine memory

Answer 7

the hypervisor

Question 8

is physical memory of a virtual OS contiguous

Answer 8

no

Question 9

MPN

Answer 9

machine page number

Question 10

how is the PPN mapped to the MPN

Answer 10

Type hypervisor keeps a page table that maps from physical to machine memory

Question 11

What is the hypervisor’s pagetable called?

Answer 11

S-PT shadow page table

Question 12

Where should the PPN->MPN be store for fully virtualized?

Answer 12

hypervisor

Question 13

Where should the PPN->MPN be store for para virtualized?

Answer 13

Guest OS

Question 14

How is VPN to MPN made efficent

Answer 14

The hypervisor traps guest PT/TLB updates of the virtualized OS. It then installs that mapping in the S-PT and hardware TLB.

Question 15

does a para guest know about the S-PT

Answer 15

The PPN to MPN can be pushed in the guest OS

Question 16

What methods does the para guest use to work with hypervisor MMU

Answer 16

CreatePT, SwitchPT, UpdatePT

Question 17

Ballooning requires what of the guest OS?

Answer 17

Needs to have a “balloon” device driver.

Question 18

How does the Ballooning process work?

Answer 18

The balloon driver keeps allocating memory until it has enough to return a page to the hypervisor

Question 19

What if the host needs less memory

Answer 19

the balloon is deflated

Question 20

Which type of virtualization uses ballooning

Answer 20

Para and full

Question 21

How can pages from separate guest OSs be shared

Answer 21

They can be copied into the same MPN and then the OS is told they are copy of write (COW)

Question 22

how does content base sharing work

Answer 22

A hint frame is created that contains a hash of the pages content. The hash is stored in a hash table upon hit the full page content is compared

Question 23

What is contained in the hint frame

Answer 23

hash of content, CM#, PPN, MPN, REF# (how many matches)

Question 24

what happens on successful match in content based sharing?

Answer 24

The PPN is updated in the guest and page is marked COW.

Question 25

Do shared code pages need to be COW

Answer 25

No. Since the pages cannot be modified

Question 26

What different memory allocation policies are used by the hypervisor

Answer 26

-pure shared based approach, - working-set based, - Dynamic idle-adjusted shared approach

Question 27

What approach does vmware es6 server take

Answer 27

if you are holding idle pages it taxes you and takes 50% of the pages away

Question 28

Why does it avoid a 100% tax?

Answer 28

In case the guest has a sudden increase in need.