Mark Russinovich’s technical blog covering topics such as Windows troubleshooting, technologies and security.
Windows Vista SP1 includes a number of enhancements over the original Vista release in the areas of application compatibility, device support, power management, security and reliability. You can see a detailed list of the changes in the Notable Changes in Windows Vista Service Pack 1 whitepaper that you can download here. One of the improvements highlighted in the document is the increased performance of file copying for multiple scenarios, including local copies on the same disk, copying files from remote non-Windows Vista systems, and copying files between SP1 systems. How were these gains achieved? The answer is a complex one and lies in the changes to the file copy engine between Windows XP and Vista and further changes in SP1. Everyone copies files, so I thought it would be worth taking a break from the “Case of…” posts and dive deep into the evolution of the copy engine to show how SP1 improves its performance.
Copying a file seems like a relatively straightforward operation: open the source file, create the destination, and then read from the source and write to the destination. In reality, however, the performance of copying files is measured along the dimensions of accurate progress indication, CPU usage, memory usage, and throughput. In general, optimizing one area causes degradation in others. Further, there is semantic information not available to copy engines that could help them make better tradeoffs. For example, if they knew that you weren’t planning on accessing the target of the copy operation they could avoid caching the file’s data in memory, but if it knew that the file was going to be immediately consumed by another application, or in the case of a file server, client systems sharing the files, it would aggressively cache the data on the destination system.
File Copy in Previous Versions of Windows
In light of all the tradeoffs and imperfect information available to it, the Windows file copy engine tries to handle all scenarios well. Prior to Windows Vista, it took the straightforward approach of opening both the source and destination files in cached mode and marching sequentially through the source file reading 64KB (60KB for network copies because of an SMB1.0 protocol limit on individual read sizes) at a time and writing out the data to the destination as it went. When a file is accessed with cached I/O, as opposed to memory-mapped I/O or I/O with the no-buffering flag, the data read or written is stored in memory, at least until the Memory Manager decides that the memory should be repurposed for other uses, including caching the data of other files.
The copy engine relied on the Windows Cache Manager to perform asynchronous read-ahead, which essentially reads the source file in the background while Explorer is busy writing data to a different disk or a remote system. It also relied on the Cache Manager’s write-behind mechanism to flush the copied file’s contents from memory back to disk in a timely manner so that the memory could be quickly repurposed if necessary, and so that data loss is minimized in the face of a disk or system failure. You can see the algorithm at work in this Process Monitor trace of a 256KB file being copied on Windows XP from one directory to another with filters applied to focus on the data reads and writes:
Explorer’s first read operation at event 0 of data that’s not present in memory causes the Cache Manager to perform a non-cached I/O, which is an I/O that reads or writes data directly to the disk without caching it in memory, to fetch the data from disk at event 1, as seen in the stack trace for event 1:
In the stack trace, Explorer’s call to ReadFile is at frame 22 in its BaseCopyStream function and the Cache Manager invokes the non-cached read indirectly by touching the memory mapping of the file and causing a page fault at frame 8.
Because Explorer opens the file with the sequential-access hint (not visible in trace), the Cache Manager’s read-ahead thread, running in the System process, starts to aggressively read the file on behalf of Explorer at events 2 and 3. You can see the read-ahead functions in the stack for event 2:
You may have noticed that the read-ahead reads are initially out of order with respect to the original non-cached read caused by the first Explorer read, which can cause disk head seeks and slow performance, but Explorer stops causing non-cached I/Os when it catches up with the data already read by the Cache Manager and its reads are satisfied from memory. The Cache Manager generally stays 128KB ahead of Explorer during file copies.
At event 4 in the trace, Explorer issues the first write and then you see a sequence of interleaved reads and writes. At the end of the trace the Cache Manager’s write-behind thread, also running in the System process, flushes the target file’s data from memory to disk with non-cached writes.
Vista Improvements to File Copy
During Windows Vista development, the product team revisited the copy engine to improve it for several key scenarios. One of the biggest problems with the engine’s implementation is that for copies involving lots of data, the Cache Manager write-behind thread on the target system often can’t keep up with the rate at which data is written and cached in memory. That causes the data to fill up memory, possibly forcing other useful code and data out, and eventually, the target’s system’s memory to become a tunnel through which all the copied data flows at a rate limited by the disk.
Another problem they noted was that when copying from a remote system, the file’s contents are cached twice on the local system: once as the source file is read and a second time as the target file is written. Besides causing memory pressure on the client system for files that likely won’t be accessed again, involving the Cache Manager introduces the CPU overhead that it must perform to manage its file mappings of the source and destination files.
A limitation of the relatively small and interleaved file operations is that the SMB file system driver, the driver that implements the Windows remote file sharing protocol, doesn’t have opportunities to pipeline data across high-bandwidth, high-latency networks like WLANs. Every time the local system waits for the remote system to receive data, the data flowing across the network drains and the copy pays the latency cost as the two systems wait for the each other’s acknowledgement and next block of data.
After studying various alternatives, the team decided to implement a copy engine that tended to issue large asynchronous non-cached I/Os, addressing all the problems they had identified. With non-cached I/Os, copied file data doesn’t consume memory on the local system, hence preserving memory’s existing contents. Asynchronous large file I/Os allow for the pipelining of data across high-latency network connections, and CPU usage is decreased because the Cache Manager doesn’t have to manage its memory mappings and inefficiencies of the original Vista Cache Manager for handling large I/Os contributed to the decision to use non-cached I/Os. They couldn’t make I/Os arbitrarily large, however, because the copy engine needs to read data before writing it, and performing reads and writes concurrently is desirable, especially for copies to different disks or systems. Large I/Os also pose challenges for providing accurate time estimates to the user because there are fewer points to measure progress and update the estimate. The team did note a significant downside of non-cached I/Os, though: during a copy of many small files the disk head constantly moves around the disk, first to a source file, then to destination, back to another source, and so on.
After much analysis, benchmarking and tuning, the team implemented an algorithm that uses cached I/O for files smaller than 256KB in size. For files larger than 256KB, the engine relies on an internal matrix to determine the number and size of non-cached I/Os it will have in flight at once. The number ranges from 2 for files smaller than 2MB to 8 for files larger than 8MB. The size of the I/O is the file size for files smaller than 1MB, 1MB for files up to 2MB, and 2MB for anything larger.
To copy a file 16MB file, for example, the engine issues eight 2MB asynchronous non-cached reads of the source file, waits for the I/Os to complete, issues eight 2MB asynchronous non-cached writes of the destination, waits again for the writes to complete, and then repeats the cycle. You can see that pattern in this Process Monitor trace of a 16MB file copy from a local system to a remote one:
While this algorithm is an improvement over the previous one in many ways, it does have some drawbacks. One that occurs sporadically on network file copies is out-of-order write operations, one of which is visible in this trace of the receive side of a copy:
Note how the write operation offsets jump from 327,680 to 458,752, skipping the block at offset 393,216. That skip causes a disk head seek and forces NTFS to issue an unnecessary write operation to the skipped region to zero that part of the file, which is why there are two writes to offset 393,216. You can see NTFS calling the Cache Manager’s CcZeroData function to zero the skipped block in the stack trace for the highlighted event:
A bigger problem with using non-cached I/O is that performance can suffer in publishing scenarios. If you copy a group of files to a file share that represents the contents of a Web site for example, the Web server must read the files from disk when it first accesses them. This obviously applies to servers, but most copy operations are publishing scenarios even on client systems, because the appearance of new files causes desktop search indexing, triggers antivirus and antispyware scans, and queues Explorer to generate thumbnails for display on the parent directory’s folder icon.
Perhaps the biggest drawback of the algorithm, and the one that has caused many Vista users to complain, is that for copies involving a large group of files between 256KB and tens of MB in size, the perceived performance of the copy can be significantly worse than on Windows XP. That’s because the previous algorithm’s use of cached file I/O lets Explorer finish writing destination files to memory and dismiss the copy dialog long before the Cache Manager’s write-behind thread has actually committed the data to disk; with Vista’s non-cached implementation, Explorer is forced to wait for each write operation to complete before issuing more, and ultimately for all copied data to be on disk before indicating a copy’s completion. In Vista, Explorer also waits 12 seconds before making an estimate of the copy’s duration and the estimation algorithm is sensitive to fluctuations in the copy speed, both of which exacerbate user frustration with slower copies.
During Vista SP1’s development, the product team decided to revisit the copy engine to explore ways to improve both the real and perceived performance of copy operations for the cases that suffered in the new implementation. The biggest change they made was to go back to using cached file I/O again for all file copies, both local and remote, with one exception that I’ll describe shortly. With caching, perceived copy time and the publishing scenario both improve. However, several significant changes in both the file copy algorithm and the platform were required to address the shortcomings of cached I/O I’ve already noted.
The one case where the SP1 file copy engine doesn't use caching is for remote file copies, where it prevents the double-caching problem by leveraging support in the Windows client-side remote file system driver, Rdbss.sys. It does so by issuing a command to the driver that tells it not to cache a remote file on the local system as it is being read or written. You can see the command being issued by Explorer in the following Process Monitor capture:
Another enhancement for remote copies is the pipelined I/Os issued by the SMB2 file system driver, srv2.sys, which is new to Windows Vista and Windows Server 2008. Instead of issuing 60KB I/Os across the network like the original SMB implementation, SMB2 issues pipelined 64KB I/Os so that when it receives a large I/O from an application, it will issue multiple 64KB I/Os concurrently, allowing for the data to stream to or from the remote system with fewer latency stalls.
The copy engine also issues four initial I/Os of sizes ranging from 128KB to 1MB, depending on the size of the file being copied, which triggers the Cache Manager read-ahead thread to issue large I/Os. The platform change made in SP1 to the Cache Manager has it perform larger I/O for both read-ahead and write-behind. The larger I/Os are only possible because of work done in the original Vista I/O system to support I/Os larger than 64KB, which was the limit in previous versions of Windows. Larger I/Os also improve performance on local copies because there are fewer disk accesses and disk seeks, and it enables the Cache Manager write-behind thread to better keep up with the rate at which memory fills with copied file data. That reduces, though not necessarily eliminates, memory pressure that causes active memory contents to be discarded during a copy. Finally, for remote copies the large I/Os let the SMB2 driver use pipelining. The Cache Manager issues read I/Os that are twice the size of the I/O issued by the application, up to a maximum of 2MB on Vista and 16MB on Server 2008, and write I/Os of up to 1MB in size on Vista and up to 32MB on Server 2008.
This trace excerpt of a 16MB file copy from one SP1 system to another shows 1MB I/Os issued by Explorer and a 2MB Cache Manager read-ahead, which is distinguished by its non-cached I/O flag:
Unfortunately, the SP1 changes, while delivering consistently better performance than previous versions of Windows, can be slower than the original Vista release in a couple of specific cases. The first is when copying to or from a Server 2003 system over a slow network. The original Vista copy engine would deliver a high-speed copy, but, because of the out-of-order I/O problem I mentioned earlier, trigger pathologic behavior in the Server 2003 Cache Manager that could cause all of the server’s memory to be filled with copied file data. The SP1 copy engine changes avoid that, but because the engine issues 32KB I/Os instead of 60KB I/Os, the throughput it achieves on high-latency connections can approach half of what the original Vista release achieved.
The other case where SP1 might not perform as well as original Vista is for large file copies on the same volume. Since SP1 issues smaller I/Os, primarily to allow the rest of the system to have better access to the disk and hence better responsiveness during a copy, the number of disk head seeks between reads from the source and writes to the destination files can be higher, especially on disks that don’t avoid seeks with efficient internal queuing algorithms.
One final SP1 change worth mentioning is that Explorer makes copy duration estimates much sooner than the original Vista release and the estimation algorithm is more accurate.
File copying is not as easy as it might first appear, but the product team took feedback they got from Vista customers very seriously and spent hundreds of hours evaluating different approaches and tuning the final implementation to restore most copy scenarios to at least the performance of previous versions of Windows and drastically improve some key scenarios. The changes apply both to Explorer copies as well as to ones initiated by applications using the CopyFileEx API and you’ll see the biggest improvements over older versions of Windows when copying files on high-latency, high-bandwidth networks where the large I/Os, SMB2’s I/O pipelining, and Vista’s TCP/IP stack receive-window auto-tuning can literally deliver what would be a ten minute copy on Windows XP or Server 2003 in one minute. Pretty cool.
Apologies Eric. That was for Norman Diamond.
Thank you Norman for providing some reasoning behind this change, and a work around.
Seemed like the place to ask WHY it was done this way, and if there were any registry tweaks.
Actually, Linux does allow caches to evict things like resident code pages from memory. The defaults have changed over time and Linux has been tuned recently (in the past four years or so) to do this much less, but it will do it. Google 'swappiness' for more information.
Quoting Andrew Morton: "My point is that decreasing the tendency of the kernel to swap stuff out is wrong. You really don't want hundreds of megabytes of BloatyApp's untouched memory floating about in the machine. Get it out on the disk, use the memory for something useful."
FWIW, I have never seen Linux mess this up. It generally swaps out code that hasn't been used in ages and only in cases where the enlarged cache will actually help. It is not wrong to do this, and it is not wrong to allow the cache to be a huge fraction of physical memory. You just have to not screw it up, and it's easy to screw it up.
If you look at the source code for gnu cp, the algorithm is really to open the source file, read in chunks and write back to the destination file. Some versions take special care for sparse files (to preserve the sparse regions), but that's it. There's no Copy Engine, no modification of paging algorithms, and no optimisation for large files - the kernel is just expected to handle the workload - both memory and cpu - just like it would for any other process.
I'm not trying to start a flamewar here (I promise) - I'm just interested in the different approaches and why the Vista team decided there was a need for a Copy Engine at all.
OK SRS, I'm not trying to start a flamewar either, but since you spoke of gnu cp... I have to avoid looking at gnu source these days so that none accidentally sneaks into my work, but I can use gnu cp as a user.
Several years ago Microsoft admitted to a problem copying directories full of short filenames and long filenames. For example, suppose a directory contains these files:
really long filename.txt
and the short name of the second file is REALLY~2.TXT
If Windows Explorer copied really long filename.txt first, then in the destination directory it would get short name REALLY~1.TXT, and then when Windows Explorer copied the original REALLY~1.TXT, it would overwrite the destination's copy of really long filename.txt. The result would be one miscopy and one lost copy.
If I recall correctly, after Microsoft figured out what their problem was, they fixed Windows Explorer. Several years ago. First copy files that have a single name, such as REALLY~1.TXT in this case, and then copy files that have long filenames.
About two weeks ago I used a one-year-old copy of gnu cp, with the -r flag. Fortunately I checked the results. The target drive was missing a file. It took several hours to track down which directory was missing a file, and then a few minutes more to guess which file had been miscopied in the same directory.
Short filenames suck, but that's no excuse this time. gnu cp should have been fixed several years ago.
Vista SP1 is still slow and whatever the logic or reason, in 14 years of using Windows PCs and servers, nothing ever went wrong with the old, lightning fast file copy in use prior to Vista.
So it was a solution in search of a problem that now makes life hell for our Vista users. Need to move 100 large TIF files? Take a lunch break ... better still, take the day off.
"in 14 years of using Windows PCs and servers, nothing ever went wrong with the old, lightning fast file copy in use prior to Vista"
Wrong. And wrong for more reasons than one that was mentioned in the comment immediately preceding yours.
Of course that doesn't mean different bugs are better, it means different bugs are different, and still bugs.
Norman Diamond: My point was not that gnu cp was bug-free, but why MS saw the need for a complex engine: I don't know anyone who complains that *nix file copying is too slow using cp, but many many people have complained about Vista in this regard.
Although I agree with you that shortnames do indeed suck, I'd argue that the bug you describe is a problem with Windows shortname preservation and filename enumeration - i.e. it should've been fixed at a much lower level than explorer. Then everyone gets the benefit.
No matter how long time the programers check copy engine to make "better" in windows XP I move files of 4 or more Gb in very less time than in vista. For Example, yesterday I move an image of 3.92gb from vista to server 2003RC2 it takes 52minutes to move on.
So for me Vista copy engine (I think Most of vista) must die.
this is my personal opinion.
"Although I agree with you that shortnames do indeed suck, I'd argue that the bug you describe is a problem with Windows shortname preservation and filename enumeration - i.e. it should've been fixed at a much lower level than explorer."
It can't be. Some callers wish that enumerations would come back sorted in some order. (For simple languages you can wish for alphabetical order, for less simple languages you can wish for orderings that might have been set by the Library of Congress or various other standards committees, etc. But no matter what you wish for, there are still cases where you're not going to get it.)
Anyway, even if a lower level spec change can please some callers, it can't please all. The app is still going to need its own engine.
Also suppose I had done cp * dest instead of using the -r flag. Then the glob would deliver filenames as command line parameters in some order and cp wouldn't even do enumeration. That would lose again. In order to produce correct results, cp will need its own engine.
"It can't be. Some callers wish that enumerations would come back sorted in some order." - not the point at all. If you implement something as twisted as short filenames (and I know MS needed to do this for compatibility reasons - but a problem of their own making), then it has to work consistently and seamlessly at the lowest possible api level. And your excellent explanation of the gnu cp bug shows that it doesn't. Regardless of glob order, or sort order be it alphabetical or Library of Congress, low level file system operations between windows filesystems have to preserve short filenames consistently. Putting a hack in explorer is just bad design.
"Regardless of glob order, or sort order be it alphabetical or Library of Congress, low level file system operations between windows filesystems have to preserve short filenames consistently."
That still isn't possible.
The number of possible longnames is somewhere around
50000 to the 256th power
(this is a moderately wild estimate, based on a wild guess at the number of UTF16 codepoints and an approximation of the maximum length of a longname component in a directory in NTFS or FAT, and not compensating for the fact that a high half of a surrogate pair must always be followed by a low half, not counting names shorter than the maximum, etc.)
The number of possible shortnames depends on the OEM code page of the currently active system locale (or options to the mount command in Linux etc.), but again a wild estimate would be somewhere around
200 to the 11th power.
You're not going to get a consistent mapping from the larger set to the smaller set. The mapping that you're going to get is going to depend on the order you create the filenames in. When each filename is created, it will get a shortname that doesn't conflict with other names that exist at the time.
I'd liked to have seen some more functional enhancements to the Vista Explorer interface. In a superficial sense that is the OS to most users.
It's about giving the user just enough choice. And Windows Explorer in Vista has if anything removed the user even further from the decision process. In fact everything done since XP sounds like bug fixes and optimization.
Why not just have a bulk copy command?
Vista added no pause/resume. And if an error occurs when copying a large number of files it just cuts out. That's why most Devs still use XCopy.
What bugs me is that the Explorer progress window looks like it's only been glammed up with Vista's new theme. Why isn't there a "File Copy Manager" that pops up when I begin a transfer? Where I can add files to the copy queue when I realise I missed some files. Instead of two copy windows competing for the same resources and all I can do is still cancel. Windows 3.11 let me do that.
And what would be powerful for a Business edition OS and make it feel like one (rather than the one without Chess) is the ability to queue Explorer commands. Just simple visual way like FTP programs have. (I don't use FTP any more.) Then I press start and go and have a coffee and let it run.
I installed SP1 to my laptop machine and copy operations really speeded up. But problem is that opening of files from network drive slowed down considerably.
I have no domain, just Win 2k3 server and there network shares. I have created account to server with same name and password what i have in my laptop, so it doesn't ask anything when i access shares.
When i copy file to local drive or from local disk to share, it is almost instantenous... but when i open same file, it takes almost 20 seconds to open. First it takes about 5 seconds for program to open (for example wordpad when i'm opening txt-file). But program is unresponsive and file itself doesn't open for another 15 seconds. Size of file doesn't seem to matter almost at all.
I fired up wireshark and looked what is going on. It seems like vista is first retrieving file then waiting 5 seconds and then asking for file oleacc.dll from server. Server replies that no such file. Then vista waits again 5 seconds and then asks file rsaenh.dll. Again server responds that no such file. Again vista waits 5 seconds and then finally opens file. I had no such behavior prior SP1 - all files opened fastly in identical environment.
Can anyone help me with this problem as it is quite annoying.
rsaenh.dll - Microsoft Enhanced Cryptographic Provider Library.
I wonder why this is needed for your copy of a text file.