bulb in hand_S

Snowflake, the billion dollar unicorn

Snowflake’s Managing Director in South Asia, Geoff Soon, talks to Enterprise IT News about the billion dollar unicorn that has managed to raise a total valuation of USD12.4 billion. What makes it tick?

EITN: Please explain about what Snowflake does. What is your flagship product and your key product?

Geoff: Snowflake is a single product: The Data Cloud. We enable organisations to unify their data in one centralised location and run multiple workloads on top of it, everything from data lakes to data science and even data sharing.

Snowflake’s Data Cloud is a global network where thousands of organisations mobilise data with near-unlimited scale, concurrency, and performance. Inside the Data Cloud, organisations have a single unified view of data so they can easily discover and securely share governed data, and execute diverse analytics workloads.

The data made available in the Data Cloud comes from customers and other data providers that store or access their data via Snowflake’s platform.

A unique group of capabilities power the Data Cloud:

  • The near-unlimited scale and efficiency of a multi-cluster shared data architecture
  • The seamless interoperability of working with data across multiple public clouds as if they were one
  • Integrated security features that cannot be turned off
  • Modern data sharing, which allows any number of organisations to share and receive live data with each other nearly instantly, without having to move or copy data

EITN: What is Snowflake’s value proposition?

Geoff: Snowflake’s Data Cloud unlocks the value of data by eliminating silos within organisations and uniting data that is scattered throughout subsidiaries, business ecosystems, geographies, and public cloud providers. It enables organisations to:

  • Make better, quicker business decisions
  • Manage costs while meeting SLAs with instant, near-infinite scaling of storage and compute
  • Create a data-driven customer journey
  • Easily transform, integrate, and analyse all of their data
  • Securely share, exchange, acquire, and monetise live, governed data in near real time
  • Build and operate data applications
  • Streamline data engineering to accelerate and simplify your ETL and ELT processes.
  • Operate across different infrastructure clouds and regions as a single cloud, while satisfying industry and regional data privacy and security requirements
  • Replace single-copy, static, passive data storage with dynamic, live shared data that is virtually always up to date and supports multiple workloads

EITN: Does your solution propose to replace data warehousing and data lakes?

Geoff: Snowflake’s Data Cloud can be used to replace both legacy data warehouses and data lakes, while serving a host of other workloads including data applications, data engineering, data science, and data sharing.

 EITN: Please share about current trends in the area of solution that you are in. How are your solutions addressing them today and in the future?

Geoff: Using analytics to leverage on existing data for data-driven business decision making.

All enterprises have growing amounts of both structured and semi-structured data. The explosion in IoT data, weblog data, and sensor data create opportunities to join these different data sources together as structured data into single queries allowing deeper insight.

Big data requires infrastructure and cost models that can accommodate fluctuations in demand including compute or storage independently of each other. Snowflake adjusts to the user’s needs making us one of the most economical and flexible platforms out there. It is infinitely scalable and flexible which provides customers with the right resources at the right time.

The capabilities offered by cloud platforms give companies the ability to share data with partners, suppliers, and regulators in real-time. With Snowflake platforms, companies can efficiently access third-party data from external sources to help them enrich their own data.

As a software-as-a-service (SaaS) company, Snowflake offers innovative features like zero-copy clones for Dev/Test. This feature eliminates the need to store multiple database copies and can be provisioned instantly which speeds up testing or application development. Users can create an unlimited number of clones and IT teams can save days or weeks from provisioning database copies allowing more storage and operational costs savings.

All enterprises have growing amounts of both structured and semi-structured data. The explosion in IoT data, weblog data, and sensor data create opportunities to join these different data sources together as structured data into single queries allowing deeper insight.

By embracing secure data sharing, organisations have the potential to broaden their customer insight and understanding, which can transform products and customer offerings. Done right, concerns about privacy and security can be reliably addressed with modern, governed and secure data monetisation practices, revealing that the real competitive advantage comes from joining the Data Cloud now.

This benefits companies as it enables customers to utilise their data as a business asset and they can even generate new revenue streams with that data. By creating a network of providers that can share data with multiple consumers (including within their own organisation), organisations can access shared data from multiple providers for analytics and decision-making.

The capabilities offered by cloud platforms give companies the ability to share data with partners, suppliers, and regulators in real-time. With Snowflake platforms, companies can efficiently access third-party data from external sources to help them enrich their own data.

EITN: How do you address current cloud computing trends of hybrid cloud and multi-cloud?

Geoff: The vast majority of companies will become cloud-only as the destination, but there’s still some way to go. It won’t be too surprising if we see a hybrid world for the next five years, as it’s very difficult to completely replace organisations’ on-premises legacy infrastructure overnight. While Snowflake can and does fully support multi-cloud and hybrid cloud strategies, its fundamental purpose is to enable business-focused data use. That means the platform is the central element in organisations data-centric business, a place to build data applications, data exchanges, and pipelines.

Three major challenges exist today that make a multi-cloud strategy incomplete:

  1. Cloud silos are created as soon as data exists in a public cloud. Because each major cloud provider created a unique offering with proprietary APIs for data management, there’s no easy way to copy or share data from cloud to cloud. Making matters worse, it’s hard to find DevOps employees who have the skillset to work in multiple clouds, which often leads to separate cloud teams within an organisation (yet another silo).
  2. Cloud services work best when users are in close proximity. As a result, geography plays a role in creating data silos by region, especially for organisations that operate in multiple locations (regions, countries, and continents).
  3. Data portability is a problem for all organisations, including those that use open source technologies and open data formats. Today, there’s no easy way to lift multiple petabytes of data to change clouds, open source or otherwise.

Snowflake provides multi-cloud including AWS, Azure and GCP and we are multi-tenanted. The true benefits of a multi-cloud strategy will not materialise until data can be shared and replicated across clouds and regions. With cross-cloud capability, organisations will be able to securely share data across regions and cloud accounts while adhering to the same rules of data sharing: data exists locally in a single source where it’s accessed rather than moved. Plus, the platform will make cross-region asynchronous data replication possible without impacting the performance of accessing primary data.

Cross-cloud capability delivers the unified data management platform needed to enable secure data sharing, fully execute multi-cloud strategies and provide organisations with a single source of truth. By enabling data to move freely, cross-cloud capability delivers on the promise of multi-cloud strategies. After all, the true power of data lies in its ability to move freely and securely without borders in order to influence decision-making. With cross-cloud, global data will finally have its say.

EITN: How do you address data sovereignty which is a regulatory requirement for some industries?

Geoff: Around the globe, cyber threat actors are capitalising on the panic and fear caused by the COVID-19 pandemic by mounting cyber attacks. Since January 2020, malicious cyber activities have intensified using coronavirus themed phishing and scams. Traditional data sharing methods are especially vulnerable which highlights the need to revisit the organisations’ data exchange channels.

Outdated approaches require additional resources and expose company data to cyber-attacks and security threats. Organisations should use a platform that enables authorised members of a cloud ecosystem to tap into live, read-only versions of the data, allowing secure and efficient data exchange and governance. Snowflake simply does not access our customers’ data. We’re not looking to monetise our customers’ data, it’s their job to monetise their own data. We have a very different point of view to organisations that are looking to mine insights about you to make money from you. We make money from effectively storing and computing data.

Snowflake’s customers’ data is always encrypted. Our approach to encryption is incredibly secure, and it gives customers full control at any time. They can instantly revoke their key. We get audited by third parties to ensure that we have the right policies in place governing the way that our staff maintain the core platforms that our customers use. In addition, Snowflake has been awarded the following certifications (SOC 2 Type 2, PCI DSS compliance and HIPAA among others) for compliance that validates the level of Snowflake security required by industries and governments.