ChetanPatil Testing

Blog

  • NagpurTech – Technology. Events. Companies. People.

    NagpurTech – Technology. Events. Companies. People.

    I started NagpurTech – Technology. Events. Companies. People., an initiative to bring together all the information about technology, events, companies, people in & around Nagpur in 2015.  It has been inactive for a while. Hopefully starting today, I will be able to keep it more active.

    If you are a student in Nagpur interested in contributing to NagpurTech, please write to me. I think such open platform can be a great opportunity for students interested in mass media.

    You can follow NagpurTech on Twitter and Facebook.

  • Hardware Is Running The World

    Hardware Is Running The World

    Photo by Fancycrave on Unsplash

    In 2011, Marc Andreessen wrote widely read article Why Software Is Eating The World. Last few years saw similar trend to what he projected, but 2016 was special and showed a new trend which not many would have expected in the year 2011.

    Year 2016: Hardware is running the world.

    Hardware has always acted like a catalyst in the world of technology, where many things are done with click of button. Eventually, to process all this information and get user what he she needs, one has to have excellent set of hardware tools, right from WiFi chip, to CPU, to routers, to powerful servers. As it has happened every year since 1971, 2016 too quietly kept introducing hardware oriented products with epic performance and perfect solutions.

    Marc in his article emphasized how technology giants like Google, Amazon and others are writing zillions of code to make everyday decisions in one’s life. Which to some extent is true, but can’t be done without the help of hardware that can process all these zillions of code.

    My goal in this blog post is to emphasize the hardware developments and products that came from same technology giants (and many others) who were mainly focused on software are now moving ahead with software is eating the world and hardware is running it motto.

    Personal Mobile Devices:

    This is the most active domain out of all in computer hardware, and 2016 again showed growing importance of having good hardware for the best OS and applications. Google introduced Pixel smartphones out of the frustration of not having good say in designing hardware and deep integration of its AI services. A smartphone designed by Google allows it to make use of the all the features beneath the architecture which otherwise would have been difficult, and it also shows other Android vendors on how to make best use of Android’s ecosystem by designing something that’s minimal yet so powerful.

    It won’t be wrong to say that Apple literally owns this domain because of deep amalgamation of hardware and software. iPhone saw another version, but what caught everyone’s attention were these tiny little Apple Airpods. It is being said, that Apple’s future products will be inspired from Airpods, because they may allow Apple to have control over on the go virtual reality (VR), a promising and upcoming domain.

    Of all the mobile hardware advancements, something very unexpected happened, Samsung’s Note 7 exploded and raised a very important question on brutal and fast product development to outperform competitors at the cost of user safety. Even after this, mobile hardware development kept pace with market demand, and also saw ARM, leader in mobile processor IP, being acquired by SoftBank. Even Lenovo realized that it can’t produce the best hardware and acquired Motorola, which has wide variety of hardware running Android.

    Overall, 2016 saw growing importance of mobile hardware, which will help software developers in writing applications that can help users to be more productive. In 2017, I think Microsoft with Windows 10 and Google to Samsung with Android OS will show importance of great hardware for great software.

    Desktops:

    Ever since smartphones got traction, desktop is something which everyone keeps writing off, but there are section of people who do want it to survive, even I do, and so does Microsoft. Microsoft’s Surface Studio was what desktop community needed, it showed how designing hardware which is not only powerful, but attractive to use can revive desktop sales. This has got other companies like HP and Dell, who have strong presence in desktop market, pumped up as Microsoft’s major goal is to not be a hardware vendor, but to show other hardware vendors how designing hardware differently can make bring best out of the Windows 10 ecosystem.

    To keep pace with growing software and improved Windows version, Intel throughout the year 2016 kept introducing powerful and efficient CPUs. This perfect development in desktop domain has forced even Apple to rethink about iMacs, and this only means better days for desktop and macOS users. And with Windows 10 finding home in ARM, 2017 is going to be a roller coaster for both hardware development and the race of which chip will run the OS.

    Servers:

    Cloud computing requires the most powerful and efficient processors, and often giants like Google, Microsoft and Facebook have complained about the lack of push by Intel, the leader in this domain. To get the best out of the server grade processors, all software giants decided to take matters in hand, and created teams that work specifically on CPUs for servers with AI and deep learning mechanisms embedded for better efficiency.

    Even Nvidia took the lead and showed the future of chip for AI. On other-hand, IBM showed the importance of domain specific chips for deep learning. And, Amazon acknowledging the need for programmable hardware for its cloud customers. Again emphasizing the importance of hardware in the software eating world.

    Embedded Devices:

    Embedded domain is very different from mobile devices, these are those smart products which are resting quietly somewhere in your office or home, and are waiting of your input to get into action.

    Amazon’s Echo, Google’s Home, Microsoft’s HoloLens and Dial all showed how future hardware may look like. Off course, these are powered by very smart software, but also showed importance of designing products differently. VR is going to see more action with hardware products coming from Oculus, Samsung, Google and may be even Apple?

    What to look for in 2017:

    One thing is pretty clear, that software needs a perfect hardware and it has taken 4 decades for both hardware-software to come together.

    Windows 10 is the best Windows version ever, Linux desktops are getting neat and stable, Android OS is more secure and defacto for mobile devices, IoT is getting its own software ecosystem and VR is finding way into people’s daily life. Intel, ARM, Qualcomm, Nvidia, Microtek are introducing the perfect hardware platform required for all different ecosystem and products to survive.

    Starting 2017, it’s not going to be all about software, vendors have to make best of both. Just having random hardware assembled at Shenzhen with Android installed on it is not going to last forever. Hardware development for smart devices will have to evolve and with stabilized software-hardware architecture, more futuristic products could be introduced.

    Ultrabook will be one domain that will get a major push along with more exciting smart hardware products from Google. Autonomous cars will crave for better self-driving hardware. VR may go mainstream with drones making deliveries. May be CES 2017 will have answers.

  • The Circle Of Mobile Ecosystem

    The Circle Of Mobile Ecosystem

    Photo by Caspar Rubin on Unsplash

    On June 29, 2007, Apple introduced iPhone and with that it reinvented three markets: Mobile Hardware, Mobile Operating System and Mobile Apps, together called as Mobile Ecosystem. Almost one year after iPhone’s launch similar ecosystem was created by Google‘s Android. Since then, with combined market share of more than 95%, these two companies have destroyed almost all other Mobile OS ecosystems including Nokia’s Symbian, Microsoft’s Windows Mobile and RIM’s BlackBerry. After Nokia’s failed acquisition and rewriting millions of codes for touch devices, Microsoft is also getting its act together and is threatening Google and Apple in the mobile arena.

    It is important to understand where each of these three different mobile ecosystems are heading, and how it’s going to affect the market, where products from watches to cars are becoming smarter every year.

    Apple’s iOS:

    Apple is good at one thing: they know which hardware is required to the run software flawlessly. Using the past experience of hardware development, they have developed iPhone as the product to beat with the sleek UI and the ease with which developers can reach out to users who are always willing to pay for apps. Also, with sister products like Macbook, iMac, iPad and iWatch, it helps loyal Apple users to seamlessly move data across devices, which today is a big advantage with all the data and computation moving to cloud.

    However, the biggest problem with Apple’s iOS is the declining market share, which primarily is due to the fact that iOS can’t be used by other OEMs, and it is very unlikely that Apple will reinvent itself by pitching iOS to any OEMs. But nothing can be predicted, may be in near future Apple might find it interesting to acquire OnePlus to create a niche set of iPhones for emerging markets.

    So, it’s pretty clear that in order to increase mobile market share, Apple has to find better ways to get new users for iPhones. They tried this by launching low cost iPhone 5C, which hasn’t got much traction. Moving forward, Apple will keep introducing better iPhones for sure, but what they need to answer is how many new users can they add with each new launch, while retaining old customers?

    Google’s Android OS: 

    Google has done everything required to make Android a mass market product with the help of OEMs. Which has lead to wide adoption and also fierce competition. Since, Android development occurs internally before it’s open sourced, the time to get new versions out by OEMs differ and sometimes many Android smartphones see end of support well before time. So, there is always a question mark on whether the Android smartphone being launched, will be supported with future Android versions or not. This basic flaw has contributed to what is termed as Android fragmentation, apart from creating security loopholes due to the lack of on the go software patches.

    This frustration lead to many Android forks, with CyanogenMod being the most popular one, but unfortunately had to terminate the project. Just having a better forked version of Android is not they way to dominate market, as Google pitches Android with Google Mobile Services, the pre-installed apps that are door to Play Store and many other useful services, and creating these services from scratch is not an easy task. Amazon has done it, but it with failed Fire Phone, its app store still relies on Android.

    For sure, Android OS is going to be the leader in mobile space for long, but it has to tackle the biggest problem of fragmentation. Solution to which is yet to be proposed, but can be done by forming developer teams from different OEMs, and helping them with new OS features while it’s being developed at Google internally, similar to how Linux Kernel development occurs. This to some extent will bring down the time to provide new Android versions, and will also help Google by providing regular security patches.

    By launching Pixel phones, Google is in no way trying to be competitor to OEMs. Google’s major revenue stream is from search and advertisement and to better project services around these, it has to showcase it by having control over set of hardwares running latest Android OS. With Pixel line of devices, it can also test new products-solutions quickly and can ensure that other OEMs are bringing better hardware by setting Pixel as a benchmark.

    In future, better Android versions will keep coming, but the bigger picture is how will non-Pixel users get hold of latest versions before the new one is out. Even today, OEMs like Samsung, Motorola, HTC and others, have good hold over smartphone market as they cater to users with different needs. What they need is ability to push new features of Android as quickly as possible which will go long way in retaining users.

    Microsoft’s Windows 10 Mobile:

    Windows Mobile was the leader in touch smartphones, but with iOS and Android it just couldn’t keep pace with the expectations. But after redrawing Windows OS from scratch for both keyboard and touch based devices, it has positioned it self as a serious competitor in smartphone domain. We are yet to see Surface smartphones with Windows 10 Mobile, but it won’t be long before Microsoft has one. With Surface phone, Microsoft will have full range of products from laptops, desktops, pads, smartphones to VR, similar to Apple and Google line of products.

    But, the biggest challenge for Microsoft with Windows 10 Mobile, is the app ecosystem and lack of OEMs. Though, it has also done lot of work to help developers write apps for Windows without rewriting the code developed for Android and iOS, developers still don’t think Windows 10 Mobile as the place to push apps to. Surely, they have got their act together in terms of hardware and right software, but to get developers to push apps for Windows Store first, it needs to be more open, innovative and should have large user base. This can only be done if they can get Surface phones and other OEMs, to consider Windows 10 Mobile as a serious alternative to Android. They have done it before with stylus based smartphones, they can surely to this time around with touch based devices, considering how aggressive they have been since last couple of years.

    Windows 10 Mobile will surely appeal to enterprise, as RIM is out of the market. By leveraging productivity apps like Office, Outlook and other cloud services, Microsoft can reinvent this space. Also, Microsoft is the only company with they ability to port its OS to any device, and support it for longer duration of time irrespective of which OEM is running it. If they can do this with Windows 10 Mobile, I think it can give Google a strong competition until it finds solution to Android fragmentation.

    So, Where Is Mobile Ecosystem Heading?

    It’s a difficult question to answer, but each of the above three companies have their own set of problems to solve. Apple needs to extend iPhone user base, Android has to solve fragmentation issue, and Microsoft needs OEMs for mobile devices.

    But two things are for sure: First, software developers will keep on finding new users to whom they can reach out to via different app stores, but they need to keep bringing best ideas forward, as retaining user and getting them to install new apps is getting difficult. Second, hardware manufacturers will have to up ante to cater to people looking for better smart devices which are supported for longer period of time.

    On the other hand, open source enthusiasts will keep forking Android, and may find a way to create their ecosystem for apps. Whatever happens, it’s going to be an exciting journey for both software and hardware developers.

    People who are really serious about software should make their own hardware — Alan Kay

  • ElecPad – Find The Best Development Board

    ElecPad – Find The Best Development Board

    Launched ElecPad – Find The Best Development Board.

    It is an initiative to bring together all the information about different development boards that are vital in making any kind of hardware that is either embedded, mobile or IoT.

    Easily browse different types of hardware that can turn ideas into reality.

  • StartupList – Find Indian Startups

    StartupList – Find Indian Startups

    Launched StartupList – Find Indian Startups.

    It is an initiative to bring together all the information about startups in India. One can easily browse startups from different cities and states in India. Also, get to know what the startups do, who founded it and many other details.

  • One Nation One Identity

    Photo by UIDAI

    Over last few years there set of terms are being widely used to simplify Indian government’s service delivery the citizens. It started with One Rank, One Pension, then came One Nation, One Tax (GST) and the latest buzz One Nation, One Election. The argument provided by the Government of India (GoI) is that simplification, cost savings and fair competition it provides to both citizen and industry is the reason to adopt such single window policy.

    I agree with the pension and tax system, but not with the election part. In this article I will briefly talk about the intention and impact of Aadhaar and my take on what should be done to fix it.

    Intention

    The birth of these terminologies in India can be traced back to Aadhaar, world’s largest ID database that provides unique identity to all Indian citizens and residents. Here the GoI’s goal was to create a database, which I will call One Nation, One Identity (ONOI), to provide better services by means of using digital technologies. In short, individual links everything to his/her Aadhaar and gets easy eKYC and benefits on top of it. As of today, it’s being said that all Indians have Aadhaar i.e. wopping ~1.3 billion people.

    Aadhaar is just not a 12 digit number. It allows authorized entities to have both demographic and bio-metric data of any individual having Aadhaar. With such massive details of each and every citizen of an emerging market, the amount of wrong doing that can be done is massive.

    Impact

    As it’s true with any new technology (or framework in this case), similarly questions are raised on Aadhaar as to what will be the impact of such unique database. I am pretty confident even GoI didn’t knew that a framework designed to provide direct to home services (by both public and private), the usage will raise questions about privacy and large scale surveillance.

    I do support Aadhaar but it certainly raises questions about privacy. In short, the intention is great but the impact it’s having is not. Also there has been conflicting statement from the Aadhaar regulatory body UIDAI about whether or not Aadhaar numbers can be made public or not. In a country of ~1.3 billion people it’s difficult to make every tech-savvy and there are always companies that are willing to take advantage of it. Please note I am not getting into the argument whether or not Aadhaar number can be used against people, I am just asking whether or not it should be public. I also know that in India nobody cares about such details as everyone wants faster solution to services.

    In short, the framework is helping unintentionally raise the big question about data privacy laws that aren’t in citizens favor as of today.

    My Take

    Supreme Court of India has reserved verdict on whether Aadhaar should be ONOI. Depending on what the outcome of this case is, personally I believe in following that needs to be done for Aadhaar:

    • There should be a full Aadhaar technology framework audit by a third party.
    • All the loopholes as much possible should be fixed.
    • Strict action against those found to be leaking Aadhaar and bluffing people.
    • Framework should have bug bounty program.
    • UIDAI should join hands with search engines and other possible valid organization to curb popping up of Aadhaar leaked numbers online.
    • It shouldn’t be mandatory (this will be decided by Supreme Court of India).
  • The Blog Of Chetan Arvind Patil

    The Blog Of Chetan Arvind Patil

    In 2011, I started writing blogs at www.ChetanPatil.info. It taught me not only about blogging but also helped me learn few web technologies and in process allowed me to connect to people from different parts of the world. I haven’t been actively blogging as I had decided when I started www.ChetanPatil.info.

    To give a new dimension to my writing, I am starting blog at #chetanpatil – Chetan Arvind Patil – chetanpatil.info/. The only goal of this blog will be to write anything that embraces hardware, software, semiconductor and technology.

  • DyPO: Dynamic Pareto-Optimal Configuration Selection for Heterogeneous MpSoCs

    Modern multiprocessor systems-on-chip (MpSoCs) offer tremendous power and performance optimization opportunities by tuning thousands of potential voltage, frequency and core configurations. As the workload phases change at runtime, different configurations may become optimal with respect to power, performance or other metrics. Identifying the optimal configuration at runtime is infeasible due to the large number of workloads and configurations. This paper proposes a novel methodology that can find the Pareto-optimal configurations at runtime as a function of the workload. To achieve this, we perform an extensive offline characterization to find classifiers that map performance counters to optimal configurations. Then, we use these classifiers and performance counters at runtime to choose Pareto-optimal configurations. We evaluate the proposed methodology by maximizing the performance per watt for 18 single- and multi-threaded applications. Our experiments demonstrate an average increase of 93%, 81% and 6% in performance per watt compared to the interactive, ondemand and powersave governors, respectively.

  • Stay Connected To Everything

    Published By: Pune Times Mirror
    Date: 15th March 2015
    Media Type: Online Media Website