Recently, I had lunch with a good friend of mine, a security professional in the retail industry who focuses on managing her organization’s security program from traditional on-premises solutions to their growing base of cloud security solutions. Like many other enterprises, they are also actively building and deploying new, in-the-cloud versions of their own in-house applications. While we tried to shift the conversation away from work, it was clearly on both our minds with the constantly growing deluge of breaches, ransomware exploits, and malware incidents at institutions large and small alike.
Of particular concern was the daunting task of integrating disparate security tools. Today’s enterprises deploy a wide variety of security solutions; from endpoint protection to malware detectors, there’s an endless variety of tools and jargon to decipher and security professionals face a major challenge in tying them all to a coherent system.
Security in the Cloud
Traditionally, Security Incident and Event Management (SIEM) and similar tools are used to provide correlation and visibility across the on-premises security solutions. However, the accelerating adoption of cloud workloads is disrupting this status quo. The cloud brings about an era of hyperconnectivity that ultimately eliminates the traditional perimeter, and places much of the enterprise’s sensitive assets and data beyond the reach of the on-premises solutions already deployed. In parallel, Cloud Access Security Brokers (CASB) are providing cloud-focused security for cloud workloads, but usually do not protect on-premises assets. All this makes the challenge of integrating security tools that much harder.
Today, many companies lack the ability to not only provide visibility into security issues across multiple tools, but drive actions across those tools as well. As we were talking, my friend said something that has stuck with me. She said that because of this she not only feels like she’s running blind, but that it’s also making her feel paranoid.
This sentiment is not uncommon. But as the hyperconnectivity that comes with a move to the cloud poses an acute challenge, it also brings great opportunity.
Analyzing Common, Modern Attacks
I’m reminded of a breach investigation our team conducted recently. Several employee accounts were compromised in a broad email phishing attack that persisted for weeks after rapid initial response and mandatory password resets. In every wave of this attack, additional accounts were abused, spreading hundreds of phishing emails within minutes to users within the company, external customers, and partners. The attackers sent an email, masquerading as a login prompt from a sanctioned app that spread quickly to an indiscriminate network of recipients. The user and entity behavior patterns were significantly out of normal range, prompting an all-hands-on-deck response.
Using anonymizing proxy VPNs, the attackers masked their IP addresses and set up email filters, as well as responders, to undermine detection. By orchestrating security across cloud providers, we were able to pinpoint the location of origin to Nigeria – a region of the world where this organization did not have users or offices. We were also able to assert that no data assets were accessed in this attack, leading us to assume that the purpose of the attack was to collect large quantities of credentials to be sold on the black market.
There are various security measures that could have prevented, or at least mitigated this attack:
- Stronger authentication through multi-factor authentication (MFA): This hinders the attacker’s ability to use phished credentials to access the victim’s account.
- Geo-aware anomaly detection: As part of a user and entity behavior analytics (UEBA) solution, this can quickly alert activity logged from abnormal locations such as, in this case, Nigeria.
- Correlation and visibility across security tools: For instance, alerts from UEBA, streamed into a SIEM, can be correlated with information from the on-premises network, to further cement confidence about a true breach. If the user is logged in and working at her desk, activity logged to cloud assets from remote countries is definitely suspect.
A Step Beyond: Orchestration Versus Consolidation
There are a number of defenses to stop attackers by orchestrating security solutions. For example, going beyond correlation, connecting security solutions so that if one tool triggers an alert or produces insight about a user, it can notify the other solutions to trigger automated actions, closing ranks across authentication, user authorization, file access, email, and more.
As enterprises continue to invest in hybrid cloud strategies, they need their fragmented security solutions to work together. Not to just ship logs to a SIEM or central repository -- but to respond to each other’s outputs, bringing an orchestrated defense across EMM, MDM, CASB, SSO, as well as firewalls and malware detectors.
While many enterprises have made substantial investments in security solutions, a lack of integration and interoperability bars organizations from gaining the full value of the available insights. As security intelligence is fragmented across multiple solutions, security teams find themselves at a disadvantage, unable to efficiently and effectively correlate information and remediate risk across infrastructure. This challenge compounds, leaving organizations unequipped to deal with the contemporary threat landscape, as they continue to invest in hybrid cloud strategies.
Organizations like my friend’s are the common case, and while workloads are shifting to the cloud, on-premises is here to stay for the foreseeable future. I predict orchestration, rather than consolidation, will be the way for enterprises to cope with security solution fragmentation. As such, orchestration will be a crucial component in dealing with evolving cyber security threats.