The convoy phenomenon

[amin...@gmail.com]

Hello,


I have just read the following IBM Research Report about Locks and convoying:

The convoy phenomenon

https://blog.acolyer.org/2019/07/01/the-convoy-phenomenon/

And i think that it is not so smart, because i am a white arab that is smart like a genius :), and i have invented a the Holy Grail of Locks that is more powerful than the above, it is a scalable Fast Mutex
that is faster than the scalable MCS Lock, read about it in my
following thoughts:

I have invented a scalable algorithm that is a scalable fast Mutex that is remarkable and that is the Holy Grail of scalable Locks, it has the following characteristics, read my following thoughts to understand:

About fair and unfair locking..

I have just read the following lead engineer at Amazon:

Highly contended and fair locking in Java

https://brooker.co.za/blog/2012/09/10/locking.html

So as you are noticing that you can use unfair locking that can have starvation or fair locking that is slower than unfair locking.

I think that Microsoft synchronization objects like the Windows critical section uses unfair locking, but they still can have starvation.

But i think that this not the good way to do, because i am an inventor and i have invented a scalable Fast Mutex that is much more powerful , because with my scalable Fast Mutex you are capable to tune the "fairness" of the lock, and my Fast Mutex is capable of more than that, read about it on my following thoughts:

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 scalable Fast Mutex has the following characteristics:

1- Starvation-free
2- Tunable fairness
3- It keeps efficiently and very low its cache coherence traffic
4- Very good fast path performance
5- And it has a good preemption tolerance.
6- It is faster than scalable MCS lock
7- Not prone to convoying.

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

And you have to look here at our DelphiConcurrent and FreepascalConcurrent:

https://sites.google.com/site/scalable68/delphiconcurrent-and-freepascalconcurrent

Thank you,
Amine Moulay Ramdane.