16638409
| Frage | Antworten |
| What effect will doubling the block size in the Unix File System have on the maximum file size? | it will be about eight times as big |
| "safe state" means that | - No deadlock has occured - No deadlock can occur |
| Page table entry usually will contain | - when the page is in memory, information where it is stored - a valid or present bit |
| A memory management unit is responsible for | managing the access to the ram (transmission of virtual memory address to physical address) |
| During a context switch, the operating system | Loads the start of the page table of the new process into a register |
| If the virtual address spave has the size 2^n and the pagesize is 2^m, the top x Bits of the virtual address contains the page number while the lower y bits contain the offset. | x=n-m, y=m |
| What consequences does increasing the page size have? | - smaller page tables - more internal fragmentation |
| In contrast to a single-level page table, multilevel page tables | typically use much less memory per process |
| The following elements are part of a virtual address | - the page number - the offset |
| Monitoring page fault frequency of a process allows us to | manage page frame allocation per process |
| Virtual Memory | needs hardware to do the memory address translation efficiently |
| With Paravirtualization, the operating system | is fully aware of being virtualized |
| A shadow page table | is not necessary when extended page tables (EPT) can be used |
| A TLB is a mechanism to | speed up the translation of virtual to physical memory |
| For a deadlock to occur, the following conditions have to be fulfilled | - Each process is either currently assigned to exact one process or is available - Processes currently holding resources that were granted earlier can request new resources - Resources previously granted cannot be forcible taken away from a process - There must be a circular list of two or more processes, each of which is waiting for a resource held by the next member of the chain |
| Operating systems NOT using the optimal pra | - Windows 95/98/2000/NT - Solaris - Linux - OSX |
| CPU utilization tends to be higher when | - There are more processes in memory - There is a higher degree of multiprogramming |
| A typical page table entry contains | - A dirty bit - A present bit - Indication if one can write to the page (The number of the page frame this page will be loaded into) |
| A memory management unit is responsible for | o Managing the user space memory o Managing the kernel space memory o Managing the user and the kernel space memory x NONE of these things is correct |
| This term describe the state of a process |
Image:
1 (binary/octet-stream)
|
| The return value of a successful exec()-system call is | o always 0 o always > 0 o always < 0 o never < 0 x NONE of these answers make sense |
| The return value of a successful fork()-system call is | Different for the parent and the child process: On success, the PID of the child process is returned in the parent, and 0 is returned in the child. On failure, -1 is returned in the parent, no child process is created. |
| A critical section is | a piece of code which, when concurrently executed by multiple threads, may produce wrong results |
| A thread is in blocked state if it | waits for an event like I/O |
| A virtual address (IA32) contains | page frame number page table number page number page size offset ) |
| A logical address of 0xA3213456 on a 32-bit page-base virtual memory system with 4 kB pages has the following offset | 0x456 4 kB = 2^12 Bytes 12 -> last 3 hex numbers |
| Which component of a process is not shared across threads? | o Register values x Heap memory o Global variables o Program memory o Stack |
| Which cannot be a valid page size? | o 1024 Bytes o 2048 Bytes x 3072 Bytes o 4096 Bytes x 130000 Bytes |
| Virtual memory | needs hardware to do the memory address translation efficiently |
| A system uses 32 bit logical addresses, a 16 kB pafe size and 36 bit physical address. What is the size of the page table? | 2^18 entries 16 kB = 2^14 -> 32 bit - 14 bit = 18 bit |
| "Safe State" menas that | - No deadlock has occured - No deadlock can occur |
| A system uses 32 bit logical addresses, a 32 kB pafe size and 36 bit physical address. What is the size of the page table? | 2^17 entries |
| A major problem with the base & limit approach to the memory translation is | It requires the process to occupy contiguous memory locations |
| Page-based virtual memory is subject to | Internal fragmentation |
| We have a paging system on a 16-bit processor with two-level page table scheme with 32 entries in the top-level page table and 32 entries in the second-level page table. What is the size of each page? | 64 Bytes 32 = 2^5 -> 16 bit - 5 bit - 5 bit = 6 bit 2^6 = 64 |
| The ideal page replacement algorithm would evict (respecting that the "optimal" algorithm is not implementable) | Any page that has not been used since the last periodic interrupt |
| In a conventional paging system, a page table entry (PTE) will not contain | - A logical page number - An illegal-page bit |
| Unlike full data journaling, ordered journaling | x Improves performance by not writing file data into the journal |
| A shadow page table | Has to be managed by the virtual machine monitor (or hypervisor) |
| A guest page walk | - describes the accesses needed for translating addresses in a virtualized environment - may tape up to 25 memory accesses in a 64 bit environment |
| Is it possible to call exec() without fork()? | Yes |
| On a MIPS or x86 processor, software running in supervisor or kernel mode | may be able to execute instructions that software running in user mode can not. |
| How can you wait for the termination of a process | waitpid |
| On modern operating systems: Exiting a program without free() is a problem because | o it results in memory leakage until the system is rebooted o the system might become very slow o the system will freeze waiting for someone to call free() o it might result in kernel panic x None of the above |
| On modern operating systems: Exiting a program without free() | is not a problem, because the processes' virtual memory is released |
| Why are page sizes powers of 2? | The displacement within the page is determined by the n least significant bits of the virtual address. So the page size is 2^n |
| All of the following are on-disk filesystem data structures expect | o inodes o data blocks o extents o the superblock x the filedescriptor |
| Which of the following is not included in an inode in Linux? (Assume the FS does not allow storing data in the inode directly) | o File owner x File name o File modification time o Pointer to the first data block o Pointer to the last data block |
| Which of the following operation systems use the optimal memory paging algorithm? | o Windows 95/98 o Windows 2000/NT o Solaris (Sun UNIX) o Linux o OS X x None of them |
| In contrast to a single-level page table, multilevel page tables | Allows to keep only part of the page table in memory |
| Virtual memory |
Lets you run processes that exceed the size of memory
makes your system more predictable
Image:
1 (binary/octet-stream)
|
| What is the job of an open file system call? | x To verify the rights the process has to access the file x To record information as to which process is accessing the file x To make the connection between a process file object or data structure and the external file |
| The concept that if a location in memory (or on disk) is referenced, the probability increases that nearby locations will be referenced soon | Principle of locality |
| Pages of virtual memory are loaded into physical memory only as they are needed | Demand paging |
| Using virtual memory techniques to treat a sequentially accessed file as routine memory access | / |
| A named collection of related information recorded on secondary storage | / |
| Behavior of a process that does not have "enough" page frames allocated to it, characterized by freqiuent page faults | Thrashing |
| A region of memory residing in the address space of two or more cooperating processes | Shared memory |
| Abstract the hardware of a single computer into several different execution environments, creating the illusion that each computing environment is running its own private computer | Virtualization |
| Collection of information referenced by the process during the process time interval (t-tau, t) | working set |
| I have a programm which uses 2GB RAM at the time i call fork(). How much RAM do the two programs use together (in sum!) after fork() on a system that uses COW? | Until they don't write something, they use 2GB (plus a minimal overhead) |
| With paging, what is saved/restored on a process context switch | All CPU registers |
| Ein TLB(Translation lookaside Buffer) | cached Adressübersetzungen |
| What are the two operations that can be performed on a condition variable? | - signal - wait |
| Provide at least three resources the operating system allocates | - memory - locks - I/O devices (interrupts ?) |
| What are two general hardware instructions that can be performed atomically? | - exchange - atomic increment |
| Two reasons why doubling the page size might increase performance | - less pagefaults - smaller page tables |
| Two reasons why doubling the page size might decrease performance | - more to copy when fork() - more to copy when swapping a page |
Möchten Sie mit GoConqr kostenlos Ihre eigenen Karteikarten erstellen? Mehr erfahren.