• IIoT platform databases – How Mail.ru Cloud Solutions deals with petabytes of data coming from a multitude of devices

      Hello, my name is Andrey Sergeyev and I work as a Head of IoT Solution Development at Mail.ru Cloud Solutions. We all know there is no such thing as a universal database. Especially when the task is to build an IoT platform that would be capable of processing millions of events from various sensors in near real-time.

      Our product Mail.ru IoT Platform started as a Tarantool-based prototype. I’m going to tell you about our journey, the problems we faced and the solutions we found. I will also show you a current architecture for the modern Industrial Internet of Things platform. In this article we will look into:

      • our requirements for the database, universal solutions, and the CAP theorem
      • whether the database + application server in one approach is a silver bullet
      • the evolution of the platform and the databases used in it
      • the number of Tarantools we use and how we came to this
      Read more →
    • Bcache against Flashcache for Ceph Object Storage

        Fast SSDs are getting cheaper every year, but they are still smaller and more expensive than traditional HDD drives. But HDDs have much higher latency and are easily saturated. However, we want to achieve low latency for the storage system, and a high capacity too. There’s a well-known practice of optimizing performance for big and slow devices — caching. As most of the data on a disk is not accessed most of the time but some percentage of it is accessed frequently, we can achieve a higher quality of service by using a small cache.

        Server hardware and operating systems have a lot of caches working on different levels. Linux has a page cache for block devices, a dirent cache and an inode cache on the filesystem layer. Disks have their own cache inside. CPUs have caches. So, why not add one more persistent cache layer for a slow disk?
        Read more →
      • Why Enterprise Chat Apps isn’t built on Server-side Database like Hangouts, Slack, & Hip chat?

        One of the most significant tools for any organization to smoothen their collaborative world is only through a real-time chat application whether the conversation takes place on mobile or desktop. Hangouts, Slack and Hipchat have been in action for businesses to establish a decent conversation between their internal employees and clients right from small-scale to enterprises.

        Those big players come into play where there requires team collaboration. The big players are built on a server-side database where the messages shared from one device to another is stored in their server database. Ultimately, this results in storing a huge amount of data within the server-side database (Cloud-database).

        The consumption of cloud storage will be pretty high. The client-side database is more efficient where the messages relayed is stored in the client device. The messages will be queued to minimize the consumption of data in the server.
        Read more →
      • A Brief Comparison of the SDS Architectures for Virtualization

        • Translation

        The search for a suitable storage platform: GlusterFS vs. Ceph vs. Virtuozzo Storage

        This article outlines the key features and differences of such software-defined storage (SDS) solutions as GlusterFS, Ceph, and Virtuozzo Storage. Its goal is to help you find a suitable storage platform.


        Let’s start with GlusterFS that is often used as storage for virtual environments in open-source-based hyper-converged products with SDS. It is also offered by Red Hat alongside Ceph.
        GlusterFS employs a stack of translators, services that handle file distribution and other tasks. It also uses services like Brick that handle disks and Volume that handle pools of bricks. Next, the DHT (distributed hash table) service distributes files into groups based on hashes.
        Note: We’ll skip the sharding service due to issues related to it, which are described in linked articles.


        When a file is written onto GlusterFS storage, it is placed on a brick in one piece and copied to another brick on another server. The next file will be placed on two or more other bricks. This works well if the files are of about the same size and the volume consists of a single group of bricks. Otherwise the following issues may arise:
        Read more →
      • Tarantool Data Grid: Architecture and Features

          In 2017, we won the competition for the development of the transaction core for Alfa-Bank's investment business and started working at once. (Vladimir Drynkin, Development Team Lead for Alfa-Bank's Investment Business Transaction Core, spoke about the investment business core at HighLoad++ 2018.) This system was supposed to aggregate transaction data in different formats from various sources, unify the data, save it, and provide access to it.

          In the process of development, the system evolved and extended its functions. At some point, we realized that we created something much more than just application software designed for a well-defined scope of tasks: we created a system for building distributed applications with persistent storage. Our experience served as a basis for the new product, Tarantool Data Grid (TDG).

          I want to talk about TDG architecture and the solutions that we worked out during the development. I will introduce the basic functions and show how our product could become the basis for building turnkey solutions.
          Read more →
        • Quintet instead of Byte — data storage and retrieval approach

            Quintet is a way to present atomic pieces of data indicating their role in the business area. Quintets can describe any item, while each of them contains complete information about itself and its relations to other quintets. Such description does not depend on the platform used. Its objective is to simplify the storage of data and to improve the visibility of their presentation.

            We will discuss an approach to storing and processing information and share some thoughts on creating a development platform in this new paradigm. What for? To develop faster and in shorter iterations: sketch your project, make sure it is what you thought of, refine it, and then keep refining the result.

            The quintet has properties: type, value, parent, and order among the peers. Thus, there are 5 components including the identifier. This is the simplest universal form to record information, a new standard that could potentially fit any programming demands. Quintets are stored in the file system of the unified structure, in a continuous homogeneous indexed bulk of data. The quintet data model — a data model that describes any data structure as a single interconnected list of basic types and terms based on them (metadata), as well as instances of objects stored according to this metadata (data).
            Read more →
          • Bypassing LinkedIn Search Limit by Playing With API

            • Translation
            [Because my extension got a lot of attention from the foreign audience, I translated my original article into English].


            Being a top-rated professional network, LinkedIn, unfortunately, for free accounts, has such a limitation as Commercial Use Limit (CUL). Most likely, you, same as me until recently, have never encountered and never heard about this thing.


            The point of the CUL is that when you search people outside your connections/network too often, your search results will be limited with only 3 profiles showing instead of 1000 (100 pages with 10 profiles per page by default). How ‘often’ is measured nobody knows, there are no precise metrics; the algorithm decides it based on your actions – how frequently you’ve been searching and how many connections you’ve been adding. The free CUL resets at midnight PST on the 1st of each calendar month, and you get your 1000 search results again, for who knows how long. Of course, Premium accounts have no such limit in place.

            However, not so long ago, I’ve started messing around with LinkedIn search for some pet-project, and suddenly got stuck with this CUL. Obviously, I didn’t like it that much; after all, I haven’t been using the search for any commercial purposes. So, my first thought was to explore this limit and try to bypass it.

            [Important clarification — all source materials in this article are presented solely for informational and educational purposes. The author doesn't encourage their use for commercial purposes.]
            Read more →
          • Ads
            AdBlock has stolen the banner, but banners are not teeth — they will be back