Hello..
About Hardware Transactional Memory and my invention that is my powerful
Fast Mutex:
"As someone who has used TSX to optimize synchronization primitives, you
can expect to see a ~15-20% performance increase, if (big if) your
program is heavy on disjoint data access, i.e. a lock is needed for correctness, but conflicts are rare in practice. If you have a lot of
threads frequently writing the same cache lines, you are probably going
to see worse performance with TSX as opposed to traditional locking. It
helps to think about TSX as transparently performing optimistic
concurrency control, which is actually pretty much how it is implemented
under the hood."
Read more here:
https://news.ycombinator.com/item?id=8169697
So as you are noticing, HTM (hardware transactional memory) and TM can
not replace locks when doing IO and for highly contended critical
sections, this is why i have invented my following powerful Fast Mutex:
More about research and software development..
I have just looked at the following new video:
Why is coding so hard...
https://www.youtube.com/watch?v=TAAXwrgd1U8
I am understanding this video, but i have to explain my work:
I am not like this techlead in the video above, because i am also an
"inventor" that has invented many scalable algorithms and there
implementions, i am also inventing effective abstractions, i give you an example:
Read the following of the senior research scientist that is called Dave
Dice:
Preemption tolerant MCS locks
https://blogs.oracle.com/dave/preemption-tolerant-mcs-locks
As you are noticing he is trying to invent a new lock that is preemption tolerant, but his lock lacks some important characteristics, this is why
i have just invented a new Fast Mutex that is adaptative and that is
much much better and i think mine is the "best", and i think you will
not find it anywhere, my new Fast Mutex has the following characteristics:
1- Starvation-free
2- Good fairness
3- It keeps efficiently and very low the cache coherence traffic
4- Very good fast path performance (it has the same performance as the
scalable MCS lock when there is contention.)
5- And it has a decent preemption tolerance.
this is how i am an "inventor", and i have also invented other scalable algorithms such as a scalable reference counting with efficient support
for weak references, and i have invented a fully scalable Threadpool,
and i have also invented a Fully scalable FIFO queue, and i have also
invented other scalable algorithms and there inmplementations, and i
think i will sell some of them to Microsoft or to
Google or Embarcadero or such software companies.
And about composability of lock-based systems now:
Design your systems to be composable. Among the more galling claims of
the detractors of lock-based systems is the notion that they are somehow uncomposable:
“Locks and condition variables do not support modular programming,”
reads one typically brazen claim, “building large programs by gluing
together smaller programs[:] locks make this impossible.”9 The claim, of course, is incorrect. For evidence one need only point at the
composition of lock-based systems such as databases and operating
systems into larger systems that remain entirely unaware of lower-level locking.
There are two ways to make lock-based systems completely composable, and
each has its own place. First (and most obviously), one can make locking entirely internal to the subsystem. For example, in concurrent operating systems, control never returns to user level with in-kernel locks held;
the locks used to implement the system itself are entirely behind the
system call interface that constitutes the interface to the system. More generally, this model can work whenever a crisp interface exists between software components: as long as control flow is never returned to the
caller with locks held, the subsystem will remain composable.
Second (and perhaps counterintuitively), one can achieve concurrency and composability by having no locks whatsoever. In this case, there must be
no global subsystem state—subsystem state must be captured in
per-instance state, and it must be up to consumers of the subsystem to
assure that they do not access their instance in parallel. By leaving
locking up to the client of the subsystem, the subsystem itself can be
used concurrently by different subsystems and in different contexts. A
concrete example of this is the AVL tree implementation used extensively
in the Solaris kernel. As with any balanced binary tree, the
implementation is sufficiently complex to merit componentization, but by
not having any global state, the implementation may be used concurrently
by disjoint subsystems—the only constraint is that manipulation of a
single AVL tree instance must be serialized.
Read more here:
https://queue.acm.org/detail.cfm?id=1454462
Thank you,
Amine Moulat Ramdane.
--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)