Simplifying Data Science with Iguazio and MongoDB: Modernization with Machine Learning

Gabriela Preiss
December 2, 2020 | Updated: May 6, 2021
#partners#technology#modernization

For the most innovative, forward-thinking companies, “data” has become synonymous with “big data” — and “big data” has become synonymous with “machine learning and AI.” The amount of data you have is raw knowledge. The ability to connect the dots in a cohesive picture that lets you see major projections, personalizations, security breaches, etc., in real time — that’s wisdom.

Or, as we like to call it, data science.

MongoDB Cloud is the leading cloud data platform for modern applications. Iguazio, initially inspired by the powerful Iguazu Falls in South America, is the leading data science platform built for production and real-time use cases. We’re both disrupting and leading various industries through innovation and highly forethought intelligence. It makes perfect sense for us to work together to create a powerful, data-driven solution.

Iguazio Data Science & MLOps platform optimizes data science for your use cases

Iguazio enables enterprises to develop, deploy, and manage their AI applications, drastically shortening the time required to create real business value with AI. Using Iguazio, organizations can build and run AI models at scale and in real time, deploy them anywhere (multi-cloud, on-prem or edge), and bring to life their most ambitious AI-driven strategies. Enterprises spanning a wide range of verticals use Iguazio to solve the complexities of machine learning operations (MLOps), and accelerate the machine learning workflow by automating the following, end-to-end processes:

  • Data collection — ingested from any diverse source, whether structured, unstructured, raw or real-time
  • Data preparation — through exploration and manipulation at scale (go big!)
  • Continuous model training — through acceleration and automation
  • Rapid model and API deployment
  • Monitoring and management of AI applications in production

As a serverless, cloud-native data science platform, Iguazio reduces the overhead and complexity of developing, deploying, and monitoring AI models, guarantees consistent and reproducible results, and allows mobilizing, scaling, and duplicating functions to multiple enforcement points.

MongoDB delivers unprecedented flexibility for real-time data science integration

With its scalable, adaptable data processing model, ability to build rich data pipelines, and capacity to scale out while doing both in parallel, MongoDB is a foundational persistence layer for data science. It allows you to use your data intelligently in complex analytics, drawing new conclusions and identifying actions to take. Data science and data analytics go hand-in-hand, fueled by big data.

The MongoDB data platform handles data analytics by:

  • Enabling scalability and distributed processing — processing data with a query framework that reduces data movement by knowing where that data is and optimizing in-place computation
  • Accelerating insights — delivering real-time insight and actions
  • Supporting a full data lifecycle — intelligent data tiering from ingestion to transactions to retirement
  • Leveraging a rich ecosystem of tools and machine learning for data science
How Iguazio and MongoDB partner to synthesize a seamless production environment
Here's a look at how Iguazio and MongoDB partner to synthesize a seamless production environment:

MongoDB and Iguazio: from research to production in weeks

Iguazio fuses with MongoDB to allow intelligent, complex data compilations that lead to real-world ML/AI results like streaming and analytics, IoT applications, conversational interfaces, and image recognition.

Data science is opening opportunities for businesses in all areas, from financial services to retail, marketing, telco, and IoT, and those opportunities create demands on data that continue to grow. Iguazio swiftly reduces the development and deployment of data science projects from months to weeks, transforming how businesses, developers, and product owners use and imagine new use cases for their data.

Together, MongoDB and Iguazio establish a joint hybrid/multi-cloud data science platform.

Iguazio and MongoDB data science platform

MongoDB’s unique features create the perfect seeding ground for Iguazio’s data science platform. They include:

  • MongoDB’s high-performing, highly ranked data platform experience
  • No data duplication
  • Optimization for real-time, an essential factor for data science
  • An elastic, flexible model that adjusts to ever-changing load requirements
  • Production that’s ready in minutes

Meanwhile, Iguazio’s powerful ML pipeline automation simplifies the complex data science layer by creating a production-ready environment with an end-to-end MLOps solution, including:

  • A feature store for managing features (online and offline) that resides in MongoDB
  • Data exploration and training at scale, using built-in distribution engines such as Dask and Horovod
  • Real-time feature engineering using Iguazio’s Nuclio-supported serverless functions
  • Model management and model monitoring, including drift detection
  • Open and integrated Python environment, including built-in libraries and Jupyter Notebook-as-a-Service

Data and data science in the real world

When we think of data, stagnant databases may come to mind. But data in action is live, quick, and moves in real time. Data science is no different — and it has quickly incorporated itself in every sector of virtually every industry:

  • Fraud prevention — distinguishing legitimate from fraudulent behavior and learning to prevent new tactics over time
  • Predictive maintenance — finding patterns to predict and prevent failures
  • Real-time recommendation engines — processing consumer data for immediate feedback
  • Process optimization — minimizing costs and improving processes and targets
  • Remote monitoring — quickly detecting anomalies, threats, or failures
  • Autonomous vehicles — continuously learning new processes and landscapes to optimize safety, performance, and maintenance
  • Smart scheduling — increasing coordination among nearly infinite variables
  • Smart mobility systems — using predictive optimization to maintain efficiency, safety, and accuracy
  • IoT & IIoT — generating insights to identify patterns and predict behavior

Data science today

MongoDB enables a more intuitive process for data management and exploration by simplifying and enriching data. Iguazio helps turn data into smarter insights by simplifying organizations’ modernization into machine learning, AI, and the ongoing spectrum of data science — and we’ve only just scratched the surface. To learn more about how, together, Iguazio and MongoDB can transform your data processes into intelligent data science, check out our joint webinar discussing multiple client use cases.

MongoDB and modernization

To learn more about MongoDB’s overall modernization strategy for moving from legacy RDBMS to MongoDB Atlas, read here.

← Previous

How the Austin Chapter of MongoDB’s Women’s Group Built Community During the Pandemic

MongoDB is on a mission to create an inclusive workplace where every single employee can thrive. With a range of established affinity groups — and new ones forming regularly — MongoDB looks for ways to amplify those groups’ efforts and help support their overall mission. When the COVID-19 pandemic forced offices to shut down and employees to work from home, our affinity groups were challenged to find creative ways to support and grow their now-remote communities. As leaders of the MongoDB Women’s Group Austin chapter, we share how we pivoted this challenge into an opportunity. First, What's the MongoDB Women's Group The MongoDB Women’s Group is a community of MongoDB employees identifying as women, nonbinary, or trans. Our mission is to create a bold, visible, and united force for gender equality. To help us get there, the MongoDB Women’s Group hosts monthly members-only meetings as well as events open to both members and allies. Relaunched in 2018, the Austin-based chapter connects women and allies in our Austin office to a community of local companies and women’s groups that can support their growth within the tech industry. Pre-COVID, we gained a lot of momentum with our events, which included a live speaker series in the office, yoga, and events focused on subjects such as fertility and imposter syndrome. When COVID-19 hit, we faced a new challenge: how do we create a sense of community for our members when everyone works completely remote? Although initially daunting, the challenge of organizing remote events was an opportunity in disguise. It enabled us to kick off a speaker series for all employees, featuring prominent women in leadership positions across the country. Enter Angie Brown, from The Home Depot. Angie was the first woman to join our remote speaker series, and we couldn’t have asked for a better person to kick it off. She began her career at The Home Depot in 1998 as an entry-level software developer and now is Vice President of Technology — Merchandising, leading a team that develops solutions to support cataloging, pricing, and assortment capabilities at the giant retail chain. She also helps to mentor aspiring leaders in a number of ways, including actively participating in Atlanta’s Women in Technology association. Here, we share some highlights from our fireside chat with Angie during which she discussed her career and provided advice on what women can do to set themselves up for success. Fireside Chat with Angie Brown MongoDB: What advice do you have for those just starting off in their careers? Angie Brown: Opportunities can look like problems and not everyone wants to run into the fire, but avoiding problems can really be a missed opportunity. That’s one important lesson I’ve learned throughout my career. Although you should have a general idea of where you want to go, you also need to be willing to flex. Things might unfold in ways you didn’t expect. If you’re too prescriptive, you might miss out on them. So, you need to find a way to strike a balance. MongoDB: You took a role in leadership fairly early. How did you change your skills and evolve as you moved up? AB: When I talk to people considering moving into management, I ask them to look at the job and determine if the required qualities and responsibilities would make them happy. It’s not just about the title and pay increase. When you pivot from being an individual contributor to being in a leadership role, servant leadership is a huge part of it. If you look at management as a way to control, you won’t be happy. If you look at it as a way to serve others and help them be successful, then you’ll find joy in that career shift. I didn’t prethink this when I first moved into management and had a little bit of an identity crisis. I was used to being the one who got things done. All of a sudden, my role and life was all about going to meetings, and I didn’t look at meetings as tangible work. I was over it. Where was the joy in this? If your joy comes from having your hands on the keyboard and needing to do things your way, then being in management would be like fitting a square peg in a round hole. At first I felt invalidated and unsure of myself because it wasn’t my hands on the keyboard. I had to work through that and do a little soul-searching. I reframed my thinking to be happy leading a team and helping them solve their problems, even if it meant I wasn’t solving them myself. I had a lightbulb moment when I moved into a director role when I realized I was still solving big problems by helping my team tackle them. There’s nothing wrong with where you find your joy and no judgement if your passion aligns as an individual contributor; we need amazing developers! Always be aware of the work itself and make sure it aligns with what you enjoy. MongoDB: How have mentors played a role in your success? AB: I wish I had invested in mentors much sooner. In the early stages of my career, I didn’t think I needed help and believed I could just figure it all out on my own. I thought asking for help was a sign of weakness. In hindsight, my mentors have absolutely formed part of who I am today. I don’t have just one mentor. Instead, I look at a topic and focus on finding a mentor for that specific topic. With that approach, I have ended up having a number of mentors. Thank you again to Angie Brown! We appreciate your insight and inspiration. If you are interested in joining MongoDB, explore our career opportunities and join an innovative team that is disrupting the database industry every day.

November 30, 2020
Next →

Data and the European Landscape: 3 Trends for 2022

The past two years have brought massive changes for IT leaders: large and complex cloud migrations; unprecedented numbers of people suddenly working, shopping and learning from home; and a burst in demand for digital-first experiences. Like everyone else, we are hoping that 2022 isn’t so disruptive (fingers crossed!), our customer conversations in Europe do lead us to believe the new year will bring new business priorities. We’re already noticing changes in conversations around vendor lock-in, thanks to the Digital Markets Act, a new enthusiasm for combining operational and analytical data to drive new insights faster, and a more strategic embrace of sustainability. Here’s how we see these trends playing out in 2022. Digital markets act draws new attention to cloud vendor lock-in in Europe We’ve heard plenty about the European Commission’s Digital Markets Act , which, in the name of ensuring fair and open digital markets, would place new restrictions on companies that are deemed to be digital “gatekeepers” in the region. That discussion will be nothing compared to the vigorous debate we expect once the EU begins the very tricky political business of determining exactly which companies will fall under the act. If the EU sets the bar for revenues, users, and market size high enough, it’s possible that the regulation will end up affecting only Facebook, Amazon, Google, Apple, and Microsoft. But a European group representing 2,500 CIOs and almost 700 organizations is now pushing to have the regulation encompass more software companies. Their main concern centers around “distorted competition” in cloud infrastructure services and a worry that companies are being locked into one cloud vendor. A trend that will likely increase in 2022 that pushes back on cloud vendor lock-in is embracing multi-cloud strategies. We should expect to see more organisations in the region pursuing multi-cloud environments as a means to improve business continuity and agility whilst being able to access best of breed services from each cloud provider. As we have always said …”it’s fine to date your cloud provider….but don’t ever marry them.” The convergence of operational and analytical data The processing of operational and analytical data is almost always contained in different data systems, each tuned to that use case and managed by separate teams. But because that data lives in separate places, it’s almost impossible for organisations to generate insights and automate actions in real time, against live data. We believe 2022 is the year we’ll see a critical mass of companies in the region make significant progress toward a convergence of their operational and analytical data. We’re already starting to see some of the principles of microservices in operational applications, such as domain ownership, be applied to analytics as well. We’re hearing about this from so many of our customers locally, who are looking at MongoDB as an application data platform that allows them to perform queries across both real-time and historical data, using a unified platform and a single query API. This results in the applications they are building becoming more intelligent and contextual to their users, while avoiding dependencies on centralized analytics teams that otherwise slow down how quickly new, data-driven experiences can be released. Sustainability drives local strategic IT choice Technology always has some environmental cost. Sometimes that’s obvious — such as the energy needs and emissions associated with Bitcoin mining. More often, though, the environmental costs are well hidden. The European Green Deal commits the European Union to reducing emissions by 55% by 2030, with a focus on sustainable industry. With the U.N. Climate Change Conference (COP26) recently completed in Glasgow, and coming off the hottest European summer on record, climate issues have become top of mind. That means our customers are increasingly looking to make their technical operations more sustainable — including in their choice of cloud provider and data centers. According to research from IDC , more than 20% of CxOs say that sustainability is now important in selecting a strategic cloud service provider, and some 29% of CxOs are including sustainability into their RFPs for cloud services. Most interesting, 26% say they are willing to switch to providers with better sustainability credentials. Historically, it’s been difficult to make a switch like that. That’s part of the reason we built MongoDB Atlas — to give our customers the flexibility to run in any region , with any of the three largest cloud providers, and to make it easy to switch between them, and even to run a single database cluster across them. Publicly available information about the footprint of individual regions and even single data centers will make it simpler for companies to make informed decisions. Already, at least one cloud platform has added indicators to regions with the lowest carbon footprint. Source: IDC, European Customers Engage Services Providers at All Stages of Their Cloud Journey, IDC Survey Spotlight, Doc #EUR248484021, Dec 2021

December 21, 2021