Exynos 7420: Multitasking, multi-core and multiprocessing

by: Gary SimsAugust 31, 2015
1.3K

Exynos 7 OctaOne of the most popular smartphone processors at the moment is the Samsung Exynos 7420, mainly because it is the processor used by Samsung for its current range of high-end devices including the Samsung Galaxy S6, the Samsung Galaxy S6+ Edge, and the Samsung Galaxy Note 5. The Exynos 7420 is an octa-core processor which means it has 8 CPU cores, each of which is capable of running a task in parallel with other tasks running on the other cores.

With 8 cores and the possibility to run 8 tasks in parallel, it is important to understand what level of parallelization is offered by this high performance CPU.

Recap

Earlier this year I wrote two in-depth articles about the nature of multiprocessing on Android and specifically on ARM based CPUs. The first article debunked the myth that Android apps only use one CPU core, while the second looked at how the Samsung Galaxy S6 uses its octa-core processor.

Both bits of research showed how Android utilizes the parallel (multi-core) nature of modern processors. Samsung’s Exynos 7420 is an ARM based processor with built-in Heterogeneous Multi-Processing (HMP). In general, the quad-core processors found  in everything from desktops to smartphones have a set of cores which are all equal in terms of their performance and power consumption. In a HMP CPU, not all the cores are equal (hence, heterogeneous). The Exynos 7420 has a cluster of Cortex-A57 cores and a cluster of Cortex-A53 cores. The A57 is a high performance core, while the A53 has greater energy efficiency. This arrangement is known as big.LITTLE, where “big” processor cores (Cortex-A57) are combined with “LITTLE” processor cores (Cortex-A53).

Perfect Dude 2: big vs LITTLE (with highlights)

Perfect Dude 2: big vs LITTLE (with highlights)

When tasks are run on the LITTLE cores they use less power, they drain the battery less, however they may run a little slower. When tasks are run on the big cores, they finish sooner but they use more battery to do so.

Once we understand that not all cores are equal, it is then interesting to see how Android uses those cores and what level of simultaneous processing occurs, and on which cores, big or LITTLE?

Workload Automation

My previous tests use a tool, which I wrote myself, to determine how the CPU is being used. It uses the various pieces of information about the activity of the Linux kernel which are made available via the /proc/stat file. However, it has a shortcoming. Since the data about the CPU usage is being generated by polling /proc/stat it means that some tasks can appear to be parallel when in fact they aren’t.

The polling interval is around one six of a second (i.e. around 160 milliseconds). If a core reports its usage is 25% in that 160 milliseconds and another core reports its usage is 25% then the graphs will show both cores running simultaneously at 25%. However it is possible that the first core ran at 25% utilization for 80 milliseconds and then the second core ran at 25% utilization for 80 milliseconds.

To delve deeper into the parallel nature of the Exynos 7420 I have switched from using my own tool to the open source Workload Automation tool. Written by ARM it is designed for running tests that exercise the CPU on Android and Linux devices. The key thing is that it supports the Linux kernel internal tracer known as ftrace.

This means that information about the exact scheduling of the CPU cores can be extracted directly from deep within the Linux kernel itself. The result of which is that the polling interval weakness of my CPU usage tool is eliminated.

Web browsing

If I was to ask you what is the most arduous task that your smartphone’s CPU performs, you might think it would be a game like Modern Combat 5 or Asphalt 8, and you would be right to a certain degree. However the thing about big 3D games is that they load the GPU just as much (or even more) than the CPU. Although the CPU is used quite heavily during 3D gaming, a big chunk of the workload is handled elsewhere. If we are looking for a job which makes the CPU sweat a bit, it is in fact web browsing!

Here is a set of graphs which show how the CPU is used when browsing the Android Authority website using Chrome:

wa-parallel-chrome-aa

 

There are three graphs. The first one on the top-left shows how the four Cortex-A53 cores are used during 90 seconds of web browsing. As you can see for 18% of the time none of the cores are being used, effectively the cluster of Cortex-A53 cores is idle. For 19% of the time 1 core is being used, for 18% of the time 2 cores are being used in parallel, 3 cores for 19%, and 4 cores for 24% of the time.

The graph on the top-right shows the same data but now for the cluster of big Cortex-A57 cores. For nearly 60% of the time one big core is in use and for 14% of the time 2 cores are in use. In fact, for over 80% of the time 1 or more Cortex-A57 cores are being used.

The graph at the bottom shows the overall level of parallelization across all of the CPU cores. For less than 4% of the time the whole CPU is idle, for 15% of the time 1 core is being used, 2 cores for 16%, and so on. What is interesting is that for over 20% of the time 5 cores are being used in parallel.

If the Exynos 7420 was a quad-core processor then the scheduler at the heart of the Linux kernel would not have the option to use 5 cores simultaneously.

If the Exynos 7420 was a quad-core processor then the scheduler at the heart of the Linux kernel would not have the option to use 5 cores simultaneously. More than that, there are moments when 6, 7 and all 8 cores of the CPU are being used in parallel.

The situation for Firefox is similar, but not the same:

wa-parallel-firefox-aa

 

As you can see, Firefox mainly uses 2 and 3 cores in parallel, however for around 10% of the time it uses more than 4 cores. For Chrome, big Cortex-A57 cores were used over 80% of the time, for Firefox that number jumps to over 90% of the time.

We shouldn't underestimate the capabilities of the Cortex-A53 cores.

At this point you might be thinking, well if Chrome and Firefox are using the big cores heavily then why not just build a CPU with just four Cortex-A57 cores and leave the Cortex-A53 cores out altogether? The answer is that the big cores use more battery life and the way big.LITTLE works is that they are only called upon when needed. The little cores are still being used for around 75% of the workload and, as we will see soon, some workloads don’t even use the big cores!

YouTube

Although we talk about big cores and LITTLE cores, we shouldn’t underestimate the capabilities of the Cortex-A53 cores. They are full 64-bit processing units which can perform exactly the same operations as the bigger Cortex-A57 cores, but they have been designed to have greater power efficiency. However for some tasks the Cortex-A53 is more than sufficient.

Here is the data captured when streaming a 720p YouTube video over Wi-Fi:

wa-parallel-youtube-aa

 

As you can see, all of the work is performed by the Cortex-A53 cores. Since the video decoding is actually performed by the GPU or a hardware video decoder, then the CPU is only responsible for the Wi-Fi, for getting the streaming data from the Internet, and for loading the right bits of memory for the video decoder to tackle the next frame. The result of this “relatively easy” load is that the big cores basically sleep the whole time. In fact, the Cortex-A53 cores spend almost one quarter of their time idle as well!

Games

So, if the YouTube app only uses the Cortex-A53 cores because a lot of the video work is done by dedicated hardware, what does that means for games? Do they use the Cortex-A57 at all? Below is three sets of graphs for three gaming apps: Asphalt 8, Epic Citadel, and Crossy Road:

If you look at these graphs you will see that there is a general pattern. For the most part the games use 1 to 3 cores of the processor and occasionally peak at using 4 or 5 cores simultaneously. The Cortex-A53 cores are used for around 60% to 70% of the time, with the cores idling for around one quarter to one third of the time. However the big cores aren’t sitting idly, as with YouTube. What we see is that for Asphalt 8 and Epic Citadel are using 1 big core for at least half the time, and that even Crossy Road tends to lean on at least one big core. This is because gaming is a more complex activity than video streaming. Lots of gaming objects to create, manipulate and track. It is likely that the active Cortex-A57 core is being used for the most complex tasks performed by the CPU and the LITTLE cores for the rest.

Other workloads

I also tested Gmail, Amazon Shopping, and Flickr. However before we look at those, I want to bring your attention to the Microsoft Word app for Android:

wa-parallel-word-aa

As you can see the Word app behaves like many other apps. It uses a mixture of the Cortex-A53 and Cortex-A57 cores and it spends a lot of the time idle, due to the nature of the app. However what is interesting is that when the app has something to do, like creating a new document, it can use all 8 CPU cores. In fact it seems that when it is busy, it jumps straight from using a couple of cores right up to 8. The amount of time it is using 5, 6, or 7 cores is much less than the time it uses 8 cores.

As for the other apps, here are their graphs for your perusal:

Wrap-up

The results of this testing is broadly in line with my previous tests and again underlines the parallel nature of Android and Android apps. It also highlights the power of Heterogeneous Multi-Processing and how the LITTLE cores are being used for most jobs and the big cores are being called upon for the heavy lifting.

Never underestimate the role of the GPU and other video hardware.

This data also shows just how powerful a processor the Exynos 7420 is. At no time is the Exynos 7420 being asked to work overly hard, and there are lots of idle moments (which are good as it means that minimal battery power is being used). That being the case, it would be interesting to see how HMP works in other combinations other than just 4+4. For example, the LG G4 uses a hexa-core processor, the Snapdragon 808, rather than an octa-core processor. The 808 uses two Cortex-A57 cores and four A53 cores. Or at the other extreme, how HMP works in the deca-core Helio X20 from MediaTek.

Finally, we must never underestimate the role of the GPU and other video hardware. Both the YouTube test and the gaming tests show the importance of the graphics part of the SoC.

So, what are your thoughts on Heterogeneous Multi-Processing, big.LITTLE, octa-core processors, hexa-core processors, deca-core processors, and the Exynos 7420? Please let me know in the comments below.

  • vampyren

    Exynos seems to be a great choice but sadly no matter how fast it is if Samsung don’t optimize their code it will not be faster then a lower spec chipset. I usually but Samsung and this time I went for LG G4 and I must say LG has done a fantastic job in optimising the code. I always get these micro delays on my Samsung phones that irritates me. I also have an iPhone and considering the dual core and memory in it I get amazed how Apple takes advantage of the hardware. On android we have these monster specs but still the phone is not performing good enough. I’m not taking about performance test here but rather real day to day usage. I wish companies could slow down the phone making ( New phone every quarter or so) and focus more on software.

    • Ben

      Yeah, software and configuration make a big difference. My S3 had okay performance on jellybean, but battery life tanked with kitkat (and performance wasn’t so hot either). but once I installed aokp battery life improved a little, and with (CM12) lollipop it was like a new phone.

      • stfu

        Microlags don’t matter at all in a practical day for a practical individual. I’d be nodding my head in agreement if you all were talking anything significant but it turns out you’re just letting out your princess complexes.

    • nebulaoperator

      So true. I wish more people realise and speak out about importance of soft optimization and hardware utilization. Sadly micro lagds do irate me too and I find it hard when picking up my next phone. I am sort of sworn android user but had iphone 6 for a few months ..it’s superbly smooth and reliable no matter ho much you load it or do…I wish ,I wish there was so great optimised soft on android… one thing! when I had M7 I felt like it’s iphone on android…

      • chris pinkston

        Totally agree. I hate micro lags , just annoys me. Stock android or a htc sense phone doesn’t have these issues. I’ve been using an old m7 lately with sense 7 ported to it. very smooth and like you said , feels like iphone on android as far as smoothness.

      • vampyren

        I said it before to others but if you really want to stay on Android for me the only option today is LG G4 specially if you care about SD card , removable battery and a top of the class camera. The code so far seems to be very well optimized so i hope LG dont scew it later. After 2 weeks of heavy use i think it performs great when multitasking. The lags are minimum and not something i actually think about where as with my Samsung it was to be expected on every single use.

      • Thejas KE

        I have been using a grand prime for a past 1 weeks as my note 5 is under repair after a hard fall and i was amazed to see a phone which costs 5 times less than note 5 performed well.The software was like stock android with a touchwiz theme.Keep in mind i did not play games in it,just used it for making calls facebook,twitter,insta and the usual web browsing.

    • Xavier_NYC

      Yeah I’ve noticed that over time and after going from Kit Kat to Lollipop my Note 4’s performance has taken a nose dive. Was getting excellent battery life and SOT but all went downhill after the update. I just like samsung phones because they always have the top of the line specs and camera but honestly might just get a nexus and call it a day.

      • Ismail Akram

        Kitkat was really fast. everyphone I’ve used when they got updated to lollipop they were not same. I had Moto X 2014 it was so fast on kitkat but lollipop made it slow. same happened with my M8 when using on kitkat its different it minor but when you start to notice you keeps noticing. can’t say much about samsung as my first smartphone was with lollipop oh no Note 4 was with kitkat. Now rocking Note 5 but I’m not please with batter life it feels sometime poor than S6 edge. I had high hope for battery when I bought Note 5 but I’m disappointed.

        • Xavier_NYC

          Yeah, the Note 4 is my first Sammy device but I’ll prob stay away from the Note 5 and S6 edge+. I thought it was silly that they made the battery smaller.

      • KenanSadhu

        Because samsung want you to change phones. It’s how they are

        • juxt417

          No people complained about the cheap plastic backs and many people also wanted wireless charging. So a complete overall is what we got. I for one love the new design and never had to change my s3’s battery in the 3 years I had that. So it doesn’t bother me at all and we shall see how these newer and hopefully better built batteries last over the next couple years.

      • vampyren

        Yupp pretty much same story here. Consider the LG G4, i just got it 2 weeks ago and its really great, never thought i would like a LG phone but man they know how to optimize the phone. Also i love having one of the best cameras and removable battery. Usually after a year i need to swap it, i kind of charge my phone several times during a day so after 1-2 year it will wear down. Today you can only go with LG G4 or Note 4 and i didnt like supporting Samsung thanks to their crappy Note 5 downgrade.

    • stfu

      Microlags don’t matter at all in a practical day for a practical individual. I’d be nodding my head in agreement if you all were talking anything significant but it turns out you’re just letting out your princess complexes. Funny.

      • Ben

        18-32hrs vs 5-10hrs is pretty significant to me. And microlags can be very irritating when trying to type or play a game and your input ends up wrong because of it.

      • vampyren

        Exactly as @disqus_HLXhQ279ow:disqus says, my S4+ is so slow now thanks to Samsung updates, when i want to make a call i even have to wait before the digits show, and the “mico” lag tend to be just lags everywhere. Really annoying so no its not just an complex, its pure garbage code optimization or lack of it on Samsungs behalf.

      • oplix

        lol angry nerd complex

        • stfu

          Try harder loser

    • godutch

      some of these delays are intentional (eg. the home button lag) but others might be too

    • tiger

      See: https://www.gamebench.net/en/blog

      All of these talks are wonderful…but in REAL LIFE, Android phones fall flat.

      Yet, Android fans accused Apple fans of falling for marketing! LOL Android fans need to look in the mirror.

      • retrospooty

        Fall flat? LOL. OK, try some more detailed benchmarks from a more thorough site . http://www.anandtech.com/show/9379/the-lg-g4-review/6

        The iPhone does OK, but not the best or fastest. Also alot of that is due to tight integration. If that is what you like, go get an iPhone, but there is also a HUGE list of missing features that goes along with that iPhone.

        • vahdyx

          All android apps run in a VM which requires more performance and typically don’t run as smooth as thing that run native. Apple has things running native which requires less performance and can run smoother than Android apps. This is why Apple must control how things are. One slip up and their entire OS can be compromised.

          Android has things running in VM which in theory can be more secure but it’s not because these VM can access what they want from the host.

          Long story short Apple will always have better app performance with less specs but that’s just the way it is. Doesn’t mean it’s better just different environments.

          • retrospooty

            Yeah, it’s a bit overly nerdy (and this is coming from a big nerd) at this point comparing benchmarks. All of these chips, even the SD810 is plenty fast. The real world speed differences are minimal. Apples is a more efficient system due to the integrated and tightly controlled nature of it, but it has alot of drawbacks as well. If someone likes a simple interface and the huge list of missing features on IOS aren’t features they use, Apple is a great platform. If someone wanted to buy it for that, its a good purchase. If you want more flexibility, and features Android wins easily. If you are buying either based on the minor differences in synthetic benchmark tests you need to have your head examined.

        • vampyren

          Should be interesting to test with the current iPhone 6S. You cant use S6 with last years phone and even using that as a standard iPhone6 performs damn well. This says a thing or two about how well and optimized the OS/software is. Maybe not as much about the hardware. I’m sure the hardware on Android phones are a beast (2-3 times better in fact) compared ti iPhone but who cares when the end result is as it is. We care about real world performance not whats on paper after all,

        • vampyren

          Great link, never thought LG G4 was this good, i know for a fact its great but this was even better then i thought. I’m really happy with this phone. Specially since it was soooo easy to add wireless charging to it :)

    • vahdyx

      You have to realize that its not necessarily Samsung’s doing. There’s a lot of people involved with android. Google, Samsung, carriers and this affects performance. I believe the international version gets 6 hours of screen on time and the carrier version hovers around 3-4 hours.

      Performance on the international version is also superior to the carrier versions as well..

    • retrospooty

      “I wish companies could slow down the phone making ( New phone every quarter or so) and focus more on software. “
      What they need to do is stop concentrating on software. Samsung, LG, HTC, can all fire 90% of their software teams and stop adding useless bloatware to their ROMS. Use stock Android and a slim skin and it will run great. They will save money and give the user a better experience. It is a shame that on OEM flagships you have to root and debloat to get the sweet performance with zero lag anywhere. Most people don’t have the tech skills to do that and to be honest, they shouldn’t have to. Dump 90% of thier software staff and keep only people that do the ROM’s with a goal of keeping it slim and trim and updating in a timely manor, which should be much easier without all the crapware they wouldn’t be putting in the way.

      • vampyren

        I have been saying this since day one :D so i totally agree with you, i worked with one of the biggest companies in the business and no one bothered to listen and they are just now starting to realize that their garbage software is just bloatware. I do love what LG has done with the G4, slim UI, usefull things here and there but its no way as heavy as Touchwiz. I do however dont think 90% is possible, there are just soooo much testing and configuring involved when making a phone, believe me i know :D

      • Ettercap

        That is a very popular opinion you hold but people forget what they would be missing without Touchwiz, not to mention many people actually like the look of Touchwiz. I can tell you because I own all of the galaxy ‘s phones up to the s6 and just bought the Note 5 that the Note 5’s Touchwiz is not the same as previous iterations they have made it extremely efficient and it is buttery smooth. A lot of your opinion is based on very old stigma it’s like if I complained about Motorola phones even though my last Motorola phone was an original Droid X circa 2009.

      • t.harris

        I buy the samsung phones because i like all the bloat ware as you call it. That why i dont by the iphone. I love the rich colours of the samsung screen it makes the iphone look dull. I can sit in my back garden and my samsung tells me that they is some one at my front door. I can sit downstairs and my samsung tells me the baby crying on the bedroom. That what i love about samsung it just so much more than a phone. I can have it to work just how l like it to my personal choice. I have the samsung galaxy edge plus . The screen is never to big on a galaxy you dont touch a samsung phone you take to it.. i had the sg1 sg2 sg4 and this sg6 edge + . I was going to try the iphone 6 + this time as its catching up to andriod but when i saw the samsung edge it made the iphone look plane. The samsung edge best looking phone in the world a class act the Berry’s the iPhone . So give me more bloat ware choise every time after all you dont have to use it or have it. Unlike the iphone you can make the samsung work how you personly want you phone. You can use it in a the simple mode that a 5 year old can use if you want to. Samsung means choice and thats what i like.

  • tiger

    See: https://www.gamebench.net/en/blog

    All of these talks are wonderful…but in REAL LIFE, Android phones fall flat…sometimes really bad!

    Yet, Android fans accused Apple fans of falling for marketing! LOL Android fans need to look in the mirror.

    • retrospooty

      smells trolly. Yup, the dumb is strong in your post – Leaves it alone.

    • Tom_Slaapstad

      Second time you’ve had to say that. Ag shame. Me thinks you doth protest too much…

  • McLaren F1P1

    The beautiful back design and mainly that beast of a processor convinced me to get the S6. I mean it’ll still be faster than any new competitor for a very long time!

    • chris pinkston

      probably not. snapdragon 820 as well as few other SOCs will probably beat the 7420 in performance and when run in a phone with more optimized and lighter software..That device will perform better than a galaxy s6. I’m excited for whatever phone uses 820 and near stock android 6.0

  • Karly Johnston

    I already debunked your methodology in the last article… the Android OS is multi-threaded and will run its background processing across several cores… the apps it runs outside of core processing does not. using your activity tracker proves nothing when you have no stable base to measure.

  • The HairStorm

    Some phone maker will introduce a 32 core processor……it will still lag……..because of all the bloatware loaded on the phone.

  • Ettercap

    Most criticism below is people who do not own a galaxy s6 or note 5 device so I’m going to throw my two cents in since I actually own both devices.

    Need longer battery life?
    Get a fast charge battery pack, mine is a 15,000 mah Aukey charges my phone 20% to 100% in around 40~45+/- minutes at full battery pack charge it can charge my phone four times over again to full.

    For the people commenting that they downgraded the battery, every test I have yet to see has shown improved battery performance with the Note 5 as Touchwiz has been scaled down big time and other energy efficient processes have been put in place.

    Need more memory?
    Get an otg microsd card reader it weighs about as much as or less than a flash drive and mine has a Keychain to attach to my car keys and can read up to 3 micro SD cards at once. It’s really not inconvenient I’m always confused when people are whining about this. They only cost around 7-10$’s.

    They also have wireless ssd’s you can connect to your phone via Bluetooth for people who can’t be troubled to plug in an otg.

    Need infrared?
    They sell cheap otg infrared devices that you can plug into your phones micro USB or alternatively your headphone jack. They also have bluetooth/nfc infrared emitters you can just stick to your phone and connect and use to transmit.

    I can guarantee that 90% of the comments below have never touched a gs6 + or note 5 as their comments seem irrelevant or very 2012 ish.