By Olaf de Senerpont Domis, senior editor at DataStax

Premji Invest is an evergreen fund formed to support the Azim Premji Foundation, which was founded by Azim Premji, the former chairman of IT services consultancy Wipro. Premji Invest deploys a “crossover format” (investing in both private and public companies) across the technology, healthcare, consumer, and FinTech landscapes; it has backed market leaders like Outreach, Sysdig, Heyday, Anaplan, Coupa, Moderna, Carta, Flipkart, Looker (acquired by Google), and DataStax. Premji Invest-US managing partner, Sandesh Patnam, established Premji Invest’s US presence in 2014 in Menlo Park, California.

We recently spoke with Sandesh to learn more about the firm’s investment strategy and his excitement about the database market.

What’s Premji Invest’s investment strategy?

We deploy a direct crossover investment strategy with a roughly equal split between mid- to late-stage growth equity and public equities. Our evergreen structure informs and supports our long-term duration investment approach: We think in 5- to 15-year horizons. While many investors view an IPO as a potential point of exit, we see the opposite. We’ve often participated meaningfully in the IPO events of standout members of our private portfolio and continued our partnership well beyond going public. We have a team in India and a team in the US, which I set up about 8-and-a-half years ago. We’re active investors and look to partner with the founders and management teams that are on a mission to create enduring companies.

What qualities do you look for in investments?

We want to invest in companies that thrive in the public markets. On the flip side, our public portfolio in many ways reflects our private practice conviction. We have deep private and public practices that operate under one hood, so it’s through this continuity that we understand the durability of a business model, pricing, quarterly cadence, value creation, and the rigor of a team. All these metrics are easier said than accomplished, but they’re a clear proxy for quality.

We also want to see significant product-market fit. I usually use the term “wild market fit.” A lot of companies can spend a lot of dollars and get a “push-based” model, but that can generate false positives. We want to see significant market “pull.” That requires some level of codification of the go-to-market strategy and process. A lot of companies have heroic sellers or unique customers — but still fail after lots of misspent dollars.

Why invest in the database market?

We’ve all heard software is going to eat the world. But more importantly, I’d say that AI [artificial intelligence] and ML [machine learning] are going to eat software. There are a lot of companies that build software that are often fairly basic workflow tools. For software to be actionable, data must be at the center. If you think of the way the world is headed with AI and ML, how is that going to get more intelligent? What is the basis of ML? In these cases, the most important aspect is: Can you organize your data and can you learn from it and piece together information in real-time that can take real action on that data?

There’s a second element that we look for: With the speed and volume at which we are accreting data, can it be stored in an efficient way? If so, your AI and ML can get better over time, so all software should be predictive in some way.

Why is the database market more interesting today than ever before?

The need for real-time information. Ecommerce, ad spend, and real-time events that happen once need to be moderated. If you don’t have the right instrumentation and analytics, provided on a real-time basis, you’re going to fall behind. 

Think about the companies that have taken share and disrupted industries; think of all the things Amazon has done and Netflix has done — all the things that the tech challengers have done to existing businesses. They all stem from the fact that they were able to instrument their business much more so than others.

At Amazon, the office of CFO is, in many ways, more like the office of the CRO [chief revenue officer]. Amazon can instrument its business at the SKU level. Think about toothpaste with a particular flavor that’s sold in the Northeast. Amazon can tell you what the profit contribution is, what the cycle of buying is, who the potential buyer is, what the supply chain logistics look like. You start breaking that down at the SKU level, and that enables the promotion of certain products in certain geographies and enables you to make specific contribution margin decisions and make near-perfect promotions.

When you think about how that’s even possible, then you start understanding the power of AWS. But you also understand that underneath AWS is the power of a very dynamic, highly distributed database. The relevance of the right data model and the right database and the impact it has on businesses is much more pronounced today. You can say that data is destiny, but I’d add that the right database is destiny — it really impacts the business model in a very profound way.

Learn more about DataStax here.

About Olaf de Senerpont Domis:


Olaf is senior editor at DataStax. He has driven awareness via content marketing roles at Google and Apigee. Prior to that, he spent two decades as a journalist, reporting on the financial services industry and technology M&A.

Data Management, IT Leadership

Modernization is on the minds of IT decision makers, and with good reason — legacy systems cannot keep up with the realities of today’s business environment. Additionally,      many organizations are discovering their modernization advantage: their developer teams, and the databases that underpin  their applications.

“Legacy modernization is really a strategic initiative that enables you to apply the latest innovations in development methodologies and technology to refresh your portfolio of applications,” says Frederic Favelin, EMEA Technical Director, Partner Presales at MongoDB.

His remarks came during an episode of  Google Cloud’s podcast series “The Principles of a Cloud Data Strategy.”

This is much more than just lift and shift,” Favelin continues. “Moving your existing application and databases to faster hardware or onto the cloud may get you slightly higher performances and marginally reduce cost, but you will fail to realize the transformational business agility and scale, or development freedom without modernizing the whole infrastructure.”      

The ‘Innovation Tax’

For many organizations, databases have proliferated, leading to a complex ecosystem of resources — cloud, on-premise, NoSQL, non-relational, traditional. The problem, Favelin says, is organizations have deployed non-relational or no-SQL databases as “band aids to compensate for the shortcomings of legacy databases.”

“So they quickly find that most non-relational databases  excel at just a few specific things — niche things — and they have really limited capabilities otherwise, such as limited queries, capabilities, or lack of data consistency,” says Favelin.

“So it’s at this point that organizations start to really feel the burden of learning, maintaining and trying to figure out how to integrate the data between a growing set of technologies. This often means that separate search technologies are added to the data infrastructure, which require teams to move and transform data from database to dedicated search engine.”

Add the need to integrate increasingly strategic mobile capabilities, and the environment  gets even more complex, quickly. In addition, as organizations are striving to deliver a richer application experience through analytics, they sometimes need to use complex extract, transform, and load (ETL) operations to move the operational data to a separate analytical database.

This adds even more time, people and money to the day-to-day operations. “So at MongoDB, we give this a name: innovation tax,” Favelin says.

Toward a modern ecosystem

Favelin says a modern database solution must address three critical needs:

It should address the fastest way to innovate, with flexibility and a       consistent developer experience. It must be highly secure, have database encryption, and be fully auditable.Next is the freedom and the flexibility to be deployed on any infrastructure– starting from laptops, moving to the cloud, and integrating with Kubernetes. It must be scalable, resilient, and mission critical with auto scaling.Finally, to offer a unified modern application experience means that the developer data platform needs to include full text search capabilities, must be operational between transactional workloads and analytical workloads, while bringing the freshness of the transactional data to the analytical data in order to be as efficient as possible to serve the best experience for the users.

“The MongoDB developer data platform helps ensure a unified developer experience,” Favelin says, “not just across different operational database workloads, but across data workloads, including search mobile data, real time analytics and more.”

Check out “The Principles of a Cloud Data Strategy”  podcast series from Google Cloud on Google podcasts, Apple podcasts, Spotify, or wherever you get your podcasts.Get started today with MongoDB Atlas on Google Cloud on Google Marketplace.

Cloud Architecture, Databases

By Aaron Ploetz, Developer Advocate

There are many statistics that link business success to application speed and responsiveness. Google tells us that a one-second delay in mobile load times can impact mobile conversions by up to 20%. And a 0.1 second improvement in load times improved retail customer engagement by 5.2%, according to a study by Deloitte.

It’s not only the whims and expectations of consumers that drive the need for real-time or near real-time responsiveness. Think of a bank’s requirement to detect and flag suspicious activity in the fleeting moments before real financial damage can happen. Or an e-tailer providing locally relevant product promotions to drive sales in a store. Real-time data is what makes all of this possible.

Let’s face it – latency is a buzz kill. The time that it takes for a database to receive a request, process the transaction, and return a response to an app can be a real detriment to an application’s success. Keeping it at acceptable levels requires an underlying data architecture that can handle the demands of globally deployed real-time applications. The open source NoSQL database Apache Cassandra®  has two defining characteristics that make it perfectly suited to meet these needs: it’s geographically distributed, and it can respond to spikes in traffic without adverse effects to its unmatched throughput and low latency.

Let’s explore what both of these mean to real-time applications and the businesses that build them.

Real-time data around the world

Even as the world has gotten smaller, exactly where your data lives still makes a difference in terms of speed and latency. When users reside in disparate geographies, supporting responsive, fast applications for all of them can be a challenge.

Say your data center is in Ireland, and you have data workloads and end users in India. Your data might pass through several routers to get to the database, and this can introduce significant latency into the time between when an application or user makes a request and the time it takes for the response to be sent back.

To reduce latency and deliver the best user experience, the data need to be as close to the end user as possible. If your users are global, this means replicating data in geographies where they reside.

Cassandra, built by Facebook in 2007, is designed as a distributed system for deployment of large numbers of nodes across multiple data centers. Key features of Cassandra’s distributed architecture are specifically tailored for deployment across multiple data centers. These features are robust and flexible enough that you can configure clusters (collections of Cassandra nodes, which are visualized as a ring) for optimal geographical distribution, for redundancy, for failover and disaster recovery, or even for creating a dedicated analytics center that’s replicated from your main data storage centers.

But even if your data is geographically distributed, you still need a database that’s designed for speed at scale.

The power of a fast, transactional database

NoSQL databases primarily evolved over the last decade as an alternative to single-instance relational database management systems (RDBMS) which had trouble keeping up with the throughput demands and sheer volume of web-scale internet traffic.

They solve scalability problems through a process known as horizontal scaling, where multiple server instances of the database are linked to each other to form a cluster.

Some NoSQL database products were also engineered with data center awareness, meaning the database is configured to logically group together certain instances to optimize the distribution of user data and workloads. Cassandra is both horizontally scalable and data-center aware. 

Cassandra’s seamless and consistent ability to scale to hundreds of terabytes, along with its exceptional performance under heavy loads, has made it a key part of the data infrastructures of companies that operate real-time applications – the kind that are expected to be extremely responsive, regardless of the scale at which they’re operating. Think of the modern applications and workloads that have to be reliable, like online banking services, or those that operate at huge, distributed scale, such as airline booking systems or popular retail apps.

Logate, an enterprise software solution provider, chose Cassandra as the data store for the applications it builds for clients, including user authentication, authorization, and accounting platforms for the telecom industry.

“From a performance point of view, with Cassandra we can now achieve tens of thousands of transactions per second with a geo-redundant set-up, which was just not possible with our previous application technology stack,” said Logate CEO and CTO Predrag Biskupovic.

Or what about Netflix? When it launched its streaming service in 2007, it used an Oracle database in a single data center. As the number of users and devices (and data) grew rapidly, the limitations on scalability and the potential for failures became a serious threat to Netflix’s success. Cassandra, with its distributed architecture, was a natural choice, and by 2013, most of Netflix’s data was housed there. Netflix still uses Cassandra today, but not only for its scalability and rock-solid reliability. Its performance is key to the streaming media company –  Cassandra runs 30 million operations per second on its most active single cluster, and 98% of the company’s streaming data is stored on Cassandra.

Cassandra has been shown to perform exceptionally well under heavy load. It can consistently show very fast throughput for writes per second on a basic commodity workstation. All of Cassandra’s desirable properties are maintained as more servers are added, without sacrificing performance.

Business decisions that need to be made in real time require high-performing data storage, wherever the principal users may be. Cassandra enables enterprises to ingest and act on that data in real time, at scale, around the world. If acting quickly on business data is where an organization needs to be, then Cassandra can help you get there.

Learn more about DataStax here.

About Aaron Ploetz:


Aaron has been a professional software developer since 1997 and has several years of experience working on and leading DevOps teams for startups and Fortune 50 enterprises.

IT Leadership, NoSQL Databases