Software as a Service has become the de-facto standard for application delivery across the enterprise. Every year, more and more desktop applications are moving to the cloud, and new native SaaS applications emerge on the market. In this model, the application code, configuration, security access control, and database now exist completely within the SaaS provider’s environment. The benefits of SaaS are many. Reduced time to value for users, lower up-front costs, ease of delivery and scalability, and continuous upgrades and new functionality. However, SaaS also presents new challenges in the realms of cybersecurity, data governance, and compliance.
Shared Responsibility in the Cloud
By now most security organizations have heard of the Shared Responsibility Model in Cloud Computing. It delineates which responsibilities fall upon the cloud provider, and which remain with the customer. The Shared Responsibility Model can be divided into 3 categories: IaaS, Paas, and SaaS.
Of these 3 categories Software as a Service (SaaS) is the most widely used, and the least understood. The Global SaaS Market size is expected to reach $185.8 billion by 2024, and rising at a compound annual growth rate of 21.4%.
Most security controls are deployed in two logical places: on the corporate network, or host operating systems. When it comes to SaaS, this approach doesn’t quite fit. The cloud provider owns both the network and the underlying operating systems that power their applications. They handle firewall and network monitoring, OS hardening, and patch management on behalf of their customers. When it comes to application security, SaaS companies are often very good at preventing common flaws such as SQL Injection and Cross Site Scripting. With all those things covered, what’s left for security teams to do?
Take another look at the Shared Responsibility Model. Customers maintain responsibility for configuring the application and its security controls, identity and access management, and data governance / compliance.
SaaS is Complex
There is a trend across enterprise IT toward slimmer endpoint operating systems. Tasks that once required bulky workstations on the desktop have moved to smartphones, tablets, Chromebooks, and IoT devices. At the same time, more and more application and business logic is moving into the cloud. Customers can now run almost any conceivable business process within a SaaS application. Software as a Service applications are flexible, powerful, and extremely customizable. They are also complex and this leaves companies at major risk of data breaches and security exposure. Not because the Cloud provider is insecure, but because the customer is unintentionally leaking their data through user error.
Today companies use a variety of tools deployed on the server and endpoint OS to help manage security. Centralized management tools for data protection, standardization, and detecting security anomalies and events are common place in the enterprise. But for SaaS applications, these OS-based tools simply don’t provide the same value. In the past, applications ran on the local OS and accessed data stored on the hard drive or across the LAN. OS based tools could effectively monitor these applications and processes and provide a layer of protection to sensitive data. With SaaS, endpoint tools only see that the browser is running. They do not have the ability to monitor the applications, processes, or data within the cloud provider’s environment.
Networks are Evolving
There are two major shifts happening in networking technology that will drastically change the way organizations protect their users and data. TLS 1.3 and the reality of 5G wireless networks.
Transport Layer Security version 1.3 is a major revision to the protocol and provides a new set of capabilities to ensure the privacy and security of network communications. As a design goal, TLS 1.3 attempts to make inspection of encrypted traffic exceedingly difficult. As browsers and cloud providers adopt support for TLS 1.3, traditional security controls that rely upon intercepting and decrypting network traffic will stop working the way they do today. Additionally, the move to 5th Generation (5G) networks will allow for data rates up to 10 Gbps. This will empower users to truly work from anywhere, and Internet of Things (IoT) devices to send and receive vast amounts of data to the cloud. The old architecture of centralizing security controls around LAN-based egress to the Internet will become less and less effective.
Security Built for SaaS
A new approach is needed to secure and monitor data access in SaaS applications. At AppOmni, we see three major gaps in security capabilities:
Help is Here
AppOmni helps security teams Scan, Secure, and Monitor their SaaS applications. Our patented technology deeply scans SaaS APIs and configurations, identifying data leaks in minutes. Apply consistent access controls and data governance across your cloud applications with our policy engine. Continuously monitor your SaaS applications for security events, and integrate detection and response capabilities with your existing workflow.