mongodb

68 results

Data in 2021: Four Predictions For an Uncertain Future

What a year it’s been. A global pandemic, a recession, and a U.S. presidential election unlike any in living memory made 2020 a tragic and tumultuous 12 months many want to forget, but can’t. Despite the uncertainty, looking back we can be sure of at least one thing: we’ve seen several years of digital disruption in a matter of months. The race to digitize as fast as possible, our “next normal”, has cut across all industries, accelerating several ancillary trends like cloud adoption, AI, and IoT. Ironically, one of the lasting effects of 2020’s profound unpredictability is just how certain we now are of the growing centrality of digitization, and therefore data, as the primary driver of business success, consumer demand, and even societal change in 2021 and beyond. As such, we asked several of MongoDB’s brightest minds to look ahead to the coming year and share their insights into how these trends in data management may play out. Petabyte-Scale Goes Mainstream The idea of “big data” isn’t new, and many firms have been working with petabyte, and even exabyte, sized data sets for some time. 2021, however, may just be the year that data finally goes “big” for everyone else. For many organizations, particularly those mid-sized and smaller, data management has until now been confined to the realm of terabytes. However, trends like the explosion of connected devices, the roll out of 5G, and the continuation of 2020’s headlong rush to digitize every aspect of business mean petabyte-scale data management is likely to become a reality for many more. And to paraphrase a famous saying: “Mo data, mo problems.” Keeping petabytes of data accessible and safe, while at the same time using it to meaningfully enrich a business, is an order of magnitude more difficult and complex than what many mid-sized enterprises are used to. Petabyte-scale data management demands stricter tolerances for uptime, scalability, and performance. In addition, the data is likely to be more distributed — on prem, in the cloud, and even across different clouds. Real-time analytics becomes a business necessity, as does taking advantage of features like automated tiering. The security and data privacy implications of holding that much data, and making it accessible to more people and connected “things,” mean petabyte-scale data management is also a business opportunity tinged with considerable financial and reputational risk. Data Privacy Continues to Be a Hot Button The coming year will further define the relationship between consumers and their data. In November, California voters approved the California Privacy Rights Act (CPRA). Along with enhancements to the already enacted CCPA (the California Consumer Privacy Act), the CPRA establishes an independent watchdog, the California Privacy Protection Agency, to enforce the CCPA now, and the CPRA when it comes into effect on January 1, 2023. There’s growing expectation that 2021 will also be the year the U.S. Federal government begins drafting a nationwide privacy law. With more states likely to follow California and enact their own CCPA-inspired privacy laws, and a new administration headed to Pennsylvania Avenue on January 20th, a national answer to the patchwork of state-based data privacy laws might finally see the light of day. An online ad for one of Apple's latest releases, a credit card Elsewhere, China and Canada are just two of several major world economies set to introduce new data privacy statutes, or overhaul existing laws over the coming 12 months. For businesses, 2021 is also set to be a landmark year for the emergence of data privacy as a competitive advantage. The latest indicator of this trend came in the closing weeks of 2020. In December, simmering tension between two of the largest and most influential companies on the planet spilled into open conflict when Facebook took out full-page advertisements in the New York Times, Wall Street Journal, and Washington Post declaring, “We’re Standing Up To Apple For Small Businesses Everywhere.” A full page ad Facebook took out in several national publications The ads were a response to changes in Apple’s iOS 14, which will prompt users to grant apps permission to gather data and track them as they move across other apps on their iPhone or iPad. That move will “break” parts of Facebook’s ad targeting system, among other things. Apple CEO Tim Cook has staked the company’s brand on becoming known as the big tech company that respects user privacy, in direct contrast to Facebook and other companies that rely heavily on customer data for their advertising-based business models. “You are not our product,” Cook said in an interview with ABC’s Diane Sawyer in 2019 . “Our products are iPhones and iPads. We treasure your data. We want to help you keep it private and keep it safe.” Make no mistake, Facebook vs. Apple is just one battle in a much larger conflict over data privacy and brand equity. No longer just a compliance challenge, the sanctity of customer data is now a business and brand burnishing advantage too. Real-time Analytics Becomes a Differentiator It’s one thing to ingest a lot of data, and quite another to put that data to use. As 2020’s digitization stampede continues, the next frontier for enterprises is to mine the information they collect for insights that drive personalized customer experiences—at scale and in real time. And to achieve this level of near-instantaneous insight and response, 2021 will be the year businesses focus their attention on moving to converged data platforms. Unlike the siloed databases of yesteryear, converged data platforms (otherwise known as translytical data platforms, like MongoDB !), combine transactional (System of Record), operational (System of Engagement), and analytical (System of Insight) workloads onto a single, unified data platform. A converged data platform allows businesses to exploit their mountains of data at the speed and efficiency consumers now demand, and all with lower complexity and risk. As business leaders seek an edge over their competition, those that prioritize real-time analytics, and move to a converged data platform, will pull further away from their peers. Not Every Cloud Has a Silver Lining From retail to recreation, hospitality to healthcare, moving data and operations to the cloud was already a right of passage on the way to digital transformation. The COVID-19 pandemic simply accelerated this move. But with speed, comes even greater risk , and embracing the cloud on an accelerated timeline is fraught with danger. Do it without proper planning—as in a simple “lift and shift” of your existing setup—and you may find the on-premise issues that currently hamper developer velocity and business agility simply follow you to the cloud. The COVID-19 pandemic has heightened the need for companies to adopt digital business models—and only cloud platforms can provide the agility, scalability, and innovation required for this transition. McKinsey, The Next-Normal Recovery Will be Digital Additionally, all the advantages the cloud affords, such as the ease of scaling your infrastructure, can quickly lead to more architectural silos and technical complexity if handled incorrectly. Our warning is that, with so many companies rushing their move to the cloud in 2021, many will fail to seize on its transformational benefits, and spend 2022 (and beyond) undoing bad architectural decisions.

December 31, 2020

MongoDB Atlas Powers Half a Billion Players of India's Favorite Mobile Pastime, Ludo King

Nothing is more human than playing games. Boards and pieces can be found from the beginnings of civilization — little scraps of technology we created to entertain ourselves. No wonder, then, that gaming is a dominant force in mobile tech. What's more surprising is that some of the most successful mobile games are versions of some of the oldest traditions. Take Ludo. A classic board game for up to four players, it can trace its direct ancestry to 6th-century India and is built from much older ideas. Players roll a die to move pieces from home along a track to a finish; the first to get all pieces there wins. You can't pass an opponent on the track, but if you land on them they go back to the start. That's it. Simple. But the way it brings players together has been enough to make Ludo the national game of the subcontinent. Now Ludo is king of the phones, in the shape of Gametion's Ludo King app. A faithful yet stylish rendition of the board game, it retains the game's simplicity and social interaction, but at an epic scale. It topped the charts for Google Play downloads in India and reached the top ten internationally, with tens of millions of players chalking up a quarter of a billion minutes of playing time a day. At one point, numbers quadrupled overnight. Yet all this was managed by a tiny team of developers who'd built their platform on MongoDB Atlas , the global cloud database service. Gametion Founder and CEO Vikash Jaiswal Ludo King's authentic board game emulation quickly tapped into the Indian psyche. "We had strong takeup right from 2016, when we launched the first version," says Gametion founder and CEO Vikash Jaiswal. "A million downloads in the first 25 days, and up to a million minutes of play a day by the start of 2020. We were doing very well already. Then came the lockdown and we went through the roof." "We Just Wanted to Concentrate on the Game" Gametion was the quintessential small gaming startup. In 2015, it had a couple of developers out of a staff of four or five, and they'd produced a suite of in-browser Flash games. The next move was obviously mobile. But at first, the company didn't move far from the idea of a simple gaming experience. Jaiswal says: "There was no database component to the Flash games, no login or user ID. We launched Ludo King in 2016 as a single player game, and soon got the user feedback that they wanted multiplayer features. You need user accounts and user data for that." The company takes pride in how quickly it can adopt and incorporate new technologies, explains Jaiswal, but that means finding the right technology to adopt. And the game was exhibiting demanding growth. "Ludo King was becoming very popular, so we knew we needed something that could scale. It had to be quick to learn — we didn't have time for complexity or long learning curves." MongoDB seemed a good fit for an underlying database. I knew it was fast and very flexible to build on, and it had lots of features. And it turned out to be a really good fit for mobile gaming — MongoDB integrates very well into our Node.js architecture. It's a native speaker. Vikash Jaiswal, Founder and CEO, Gametion Jaiswal's team was able to rely on MongoDB's flexible data model to continually expand the game's features, including more options for players and monetisation tactics. That's never stopped. In 2020, Gametion introduced two new in-game features: voice chat and egreetings to users. But they had no interest in the nuts and bolts of database administration. "We didn't want to make our own backend or worry about scaling, management or any of that. We just wanted to concentrate on the game," says Jaiswal. MongoDB Atlas hadn't made its debut yet at the time — Gametion being ahead of the game -- so the company chose the third-party mLab platform for hosting. Then in 2019, after mLab was acquired by MongoDB Inc, Gametion transitioned from mLab to MongoDB Atlas, the platform made and managed by the company behind the database. MongoDB Atlas: A 'Native Speaker' for Mobile Gaming Transitions can be challenging, but with the same underlying architecture and the support of MongoDB itself, this one was straightforward. In fact, it was so uneventful that Jaiswal says he can't remember it happening. "I don't recall any problems at all. There was no downtime, which I definitely would have remembered. MongoDB managed it all for us. The migration must have been very smooth." Once on MongoDB Atlas, running on AWS's cloud infrastructure, the team — which was now five developers — quickly found the features that mattered, such as Continuous Cloud Backup and Performance Advisor . "The dashboard is very cool. We can dial up the performance we need when we need it, and see exactly what's going on." Ludo King's Lockdown Gametion's emphasis on common open standards and a component approach has made it easy to add other functions as the game demands, maintaining a regular schedule of updates that keep the users engaged. "You can think of it as a microservices architecture. We use Kafka to manage data movement and synchronize between services. It's another way to optimize resource use across the board without sacrificing scalability or release cadence." Infrastructure Diagram for Ludo King That's something you need when you go from being one of the top mobile games in India to the uncontested champ. "At the start of March 2020, we had between 150,000 and 200,000 simultaneous users, but when lockdown hit that month, it jumped to a million, 1.5 million. We went from 8,000 IOPS to peaking at 35,000." "With 145 million downloads in the first week of lockdown alone, quickly finding the rights answers was important," says Jaiswal. "We have 50 million users a day, averaging 50 minutes of gameplay each. Some of them are on for five, six hours at a stretch." MongoDB is Integral to Future Growth The future will see more features on Ludo King, such as league tables and what Gametion sees as its primary revenue generator: in-app purchases. It'll also see some brand-new games. MongoDB is integral to this strategy, both to power innovation and to manage the consequences of success. And Gametion's roadmap is growing with its market, which means it will need features for economically managing huge numbers of casual users. " Atlas Data Lake looks useful," says Jaiswal. "We want to move inactive players — those who haven't been online in a while — away from the main database, but we don't want to just delete them." Efficiently managing hundreds of millions of users — and supporting near-instantaneous, 1,000% growth — would have once required the resources of a large corporation. But for Gametion, which still has fewer than 100 employees, these aren't limiting factors. In August 2020, India Prime Minister Narendra Modi even highlighted the success of the the game during his monthly radio programme. Ludo King is helping to fulfill the vision of popularising Indian games with a global audience. For now, Gametion's focus is growth. And MongoDB is part of that experience, the game piece that shows where you are and implements your strategy, quietly and efficiently. MongoDB Atlas is not just a database, it's a genuine game changer. Try MongoDB Atlas Free

October 9, 2020

The Top 6 Questions From AWS re:Invent 2018

Hey there, MongoDB Community! I'm Lauren Schaefer ( @Lauren_Schaefer , linkedin.com/in/laurenjanece ), MongoDB's newest developer advocate. I've only been on the job a couple of weeks, but I had the opportunity to travel to fabulous Las Vegas, Nevada, last week to speak with many of you at the MongoDB Atlas booth at AWS re:Invent 2018. The people I chatted with complimented MongoDB over and over again. I heard things like, "The performance is great!" and "When I get to choose what database I use, I choose MongoDB" and "I love Mongo!" People also asked me a lot of questions. I’ve compiled those questions into a list of the top 6 most frequently asked questions at AWS re:Invent 2018. 6. Are the socks different sizes? My primary job at the booth was to give out socks. And I gave out a LOT of socks. Several people told me that they wear the MongoDB socks they received at last year's conference all the time. I even had people show me the MongoDB socks they were wearing. Since I was giving out so many socks, one of the most common questions I received was, "Are the socks different sizes?" The socks were all one size, but they seemed to stretch to fit a variety of sizes--they’re built to scale! 5. What is Atlas? To be fair, this question probably came up so frequently because I asked people, "Are you familiar with Atlas?" as I was handing them socks to which they commonly replied, "No. What is Atlas?" You can think of MongoDB Atlas as MongoDB-as-a-service. Atlas is a fully managed, global cloud database. Atlas takes care of all the operations related to running a MongoDB database in production -- security, availability, upgrades, and patches -- so you can focus on your data and your app. You can get more details in the video below. 4. Can I see a demo of Atlas? As you can probably imagine, people were pretty excited about Atlas when they heard about it, so they wanted to see a demo. We had experts on-hand ready to give demos. For those of you who weren't able to get a demo in-person, below is a demo of how to get started with Atlas. 3. Is Atlas new? People were very excited about Atlas and many were surprised they hadn't heard of it before. A common question was, "Is Atlas new?" No, Atlas is actually a little over two years old. It was officially announced at MongoDB World on June 28, 2016. 2. What is the Atlas pricing model? Before people started to get too excited about Atlas, they wanted to know if there was a catch. They'd ask, "What's the pricing model?" Atlas has a free tier so you can tinker and begin early development without paying a thing. You don’t even need to provide credit card information to get started. Once you exceed the free tier, Atlas is billed hourly based on how much you use. Check out the Atlas Pricing page for more details on the pricing model. The Atlas Pricing page also has a pricing calculator so you can estimate how much Atlas would cost for your particular use case. 1. Why would I choose MongoDB over Amazon DynamoDB? Since we were at an Amazon conference, many people were curious about the differences between MongoDB and Amazon's DynamoDB. You can get a detailed comparison of the two on the Comparing DynamoDB and MongoDB page. Some of the key points that resonated with people at the conference were: MongoDB provides built-in document validation. Users can enforce checks on document structure, data types, data ranges, and the presence of mandatory fields. DynamoDB has limited support for different data types. As a result, developers must preserve data types on the client, which adds complexity and reduces data re-use across different applications. DynamoDB does not have native data validation capabilities. MongoDB documents can be up to 16 MB in size whereas DynamoDB items or records can be up to 400 KB in size. MongoDB provides for more flexible indexing and querying. For example, MongoDB indexes are consistent with data whereas DynamoDB indexes are sized and provisioned separately from data. Also, MongoDB allows for querying and analyzing data in multiple ways including single keys, ranges, faceted search, graph traversals, and geospatial queries. DynamoDB allows for key-value queries. MongoDB can be deployed anywhere or as a service with MongoDB Atlas on Amazon Web Services (AWS), Google Cloud Platform (GCP), or Microsoft Azure, so you are not locked into a particular vendor. DynamoDB is available as a service on AWS. Summary I had a blast meeting so many of you at AWS re:Invent, and I hope to meet many more of you at upcoming events ! Give Atlas a shot and let me know what you think!

December 3, 2018

Capture IOT Data With MongoDB Stitch in 5 Minutes

Capturing IOT data is a complex task for 2 main reasons: We have to deal with a huge amount of data so we need a rock solid architecture. While keeping a bulletproof security level. First, let’s have a look at a standard IOT capture architecture: On the left, we have our sensors. Let’s assume they can push data every second over TCP using a POST and let’s suppose we have a million of them. We need an architecture capable to handle a million queries per seconds and able to resist any kind of network or hardware failure. TCP queries need to be distributed evenly to the application servers using load balancers and finally, the application servers are able to push the data to our multiple Mongos routers from our MongoDB Sharded Cluster . As you can see, this architecture is relatively complex to install. We need to: buy and maintain a lot of servers, make security updates on a regular basis of the Operating Systems and applications, have an auto-scaling capability (reduce maintenance cost & enable automatic failover)... This kind of architecture is expensive and maintenance cost can be quite high as well. Now let’s solve this same problem with MongoDB Stitch! Once you have created a MongoDB Atlas cluster , you can attach a MongoDB Stitch application to it and then create an HTTP Service containing the following code: exports = function(payload, response) { const mongodb = context.services.get("mongodb-atlas"); const sensors = mongodb.db("stitch").collection("sensors"); var body = EJSON.parse(payload.body.text()); body.createdAt = new Date(); sensors.insertOne(body) .then(result => { response.setStatusCode(201); }); }; And that’s it! That’s all we need! Our HTTP POST service can be reached directly by the sensors from the webhook provided by MongoDB Stitch like so: curl -H "Content-Type: application/json" -d '{"temp":22.4}' https://webhooks.mongodb-stitch.com/api/client/v2.0/app/stitchtapp-abcde/service/sensors/incoming_webhook/post_sensor?secret=test Because MongoDB Stitch is capable of scaling automatically according to demand, you no longer have to take care of infrastructure or handling failovers. Next Step Thanks for taking the time to read my post. I hope you found it useful and interesting. If you are looking for a very simple way to get started with MongoDB, you can do that in just 5 clicks on our MongoDB Atlas database service in the cloud. You can also try MongoDB Stitch for free and discover how the billing works . If you want to query your data sitting in MongoDB Atlas using MongoDB Stitch, I recommend this article from Michael Lynn .

October 3, 2018