why no updates to Defraggler in a year?

[Melchior]

1 minute ago, Augeas said:

I have to beg to differ with parts of this.... the third line actually... the first half...

Wear levelling is not seen by the operating system, NTFS, or any defragging software. NTFS, or FAT, allocates a cluster number to a file which remains unchanged for the life of the file, unless some user/client action (such as a defrag) occurs. Defraggers get their fragmentation information from the MFT, not from the disk, i.e. how many datarun entries there are in the MFT record for a file. What physical pages on an SSD are actually allocated is anyone's guess.

interesting.

 

[nukecad]

Agreed, intetesting.

Guess I need to look again at wear levelling. (or maybe at where defragmenters get their information from, which of course may/will be different for different defragmenters).

Although, as I said in a different thread. I've come to the opinion that defragmenting of any drive is not realy needed (and hasn't been for a couple of decades).

If you didn't have a defragmenter showing you a drive map you would never notice any difference in performance - which is one reason why the built in Windows tool no longer shows you a drive map.

[Augeas]

Wear levelling is entirely within  the SSD and under the control of the aptly named SSD controller. If you can see what's in the Flash Translation Layer you're probably in a research lab.

If a defragger looked at a disk, looking only at allocated clusters as defined in the cluster bit map, it would just see endless clusters of meaningless data. Where does a file start, and end, and are those thousand clusters in a row all one file, or file fragments, or what? There's no way to know. So defraggers look at the MFT (ignoring FAT for the moment). On;y there is the connection between file and data clusters, and only there are file fragments defined and located.

[Himeko]

Fragmentation does not (can't) happen with SSDs in the way HDDs do, if anything it shortens its lifespan when used, you should only TRIM SSDs (Windows does this automatically or you can check under disk options/tools/optimize).

Not too sure what "SSD optimizer" apps do exactly, like Auslogics', but I guess they just trigger the TRIM command.

[Augeas]

Fragmentation on an SSD happens exactly the same as it does on an HDD. Fragmentation is a file with its data spread across multiple and separate groups of clusters, or pages, and it's the file system that determines that. It happens in the same way whether the storage device is an HDD or an SSD, although the consequences may be different.

[Melchior]

Piriform, you say you have updated Defraggler for business..?!
why no updates for general release?
please don't abandon your customers I paid once for Defraggler a long time ago..

there is no point to pay for it since you have made no updates in years... :(

[Melchior]

bumping my thread.

still no updates to Defraggler released to the general public..

I am so sad.

 

[HSchulze]

5th year, and large hard drive users like myself would like some improvements.  I would buy the Biz version if your website showed that it had more features. Performance is important for video editing, dropping to 20% on a fragmented drive.  I already use SSD pairs to bounce edits, but on a productive day, I can generate / replace 1TB or more of files.  Making the drives larger just exacerbates the problem (ie RAID of 2-5 drives, striping, etc), increases the peak speed but does not decrease the defrag time.

1) Please allow processing more than one drive at a time. Exists in competitive software.

2) Proper multi-thread support to process the file/fragment sorting operations in parallel. 

3) Increase the allowed amount of files being read/written so that scatter-gather can increase the throughput.  On 32GB machine, only 300MB used (apparently).  I see some defrag software do long read/write operations under some circumstances, much faster than reading and writing single small files.

4) Decide whether moving the next 50,000 semi-consecutive files one at a time is slower than reading them all at once, and writing them in a consolidated block is faster.  Making use of the hard drive's internal DRAM for read-ahead makes sense to speed up the software.  I know it would be fast, because on an SSD you can move 500MB/sec or more.

At >3 days per drive (70% used 12TB, not even the latest 24TB) capable of 220MB/s, it takes 10 hours to copy the drive to an empty one, and 10 hours to copy it back.  Need to do this type of defrag once a month or so.  Consolidation (full or fragmented) is important because writing new files scatters into all the potholes.  It shouldn't take an hour to do a fast consolidation.

I would put $100 on the table for 2 or more of the above.