GCP Archives | Encryption Consulting

Cloud Data Lake Security – Protecting your Cloud Data Lake

by Riley Dickens

Posted on January 8, 2021Last updated on August 18, 2022

As businesses continually move their services online to services, like Google Cloud Platform (GCP) and Amazon Web Services (AWS), the need to protect this data grows as well. This solution is referred to as Data Lake Protection. A Data Lake is one place where all of an organization’s data is stored at the same time. Storing all information in one place has a multitude of advantages. Data from different teams can be accessed all in one place and analyzed to correlate data and create better strategies. IT infrastructure also becomes much simpler to manage, as there is only one location where all data is stored. Processes from analysis to auditing are made much more streamlined as well.

Data Lakes make the majority of processes and tasks much simpler, but this means Data Lake Protection is the number one priority of organizations using these Data Lakes. The main desire when creating a secure Data Lake Protection plan is to limit access to data to only those who need access. This is called the principle of Least Privilege. The way Least Privilege works is that access to one portion of the organization is disabled for those who do not need access to it. For example, if Gary in sales wants to access the human resource records on Roger, he cannot due to the policies in place only allowing him access to sales, and sales-related, data. One way many companies create a Data Lake is by migrating their data to the cloud.

Data Lakes on the Cloud

Cloud Service Providers (CSPs) like GCP, AWS, and Microsoft Azure provide an easy and inexpensive way of creating a Data Lake for any organization’s data. By migrating IT infrastructure, like databases, from on-premises to the cloud, a Data Lake is formed. Cloud Data Lakes are becoming more and more common on the cloud, as CSPs provide a variety of helpful tools to analyze and secure data. Encryption management can be left to the CSP, or the user can control it with Hardware Security Modules, encryption key management, and Google Cloud Functions.

Best Practices

To begin the process of protecting an organization’s Data Lake, there are best practices one should follow. These best practices are:

Principle of Least PrivilegeAs previously noted, the principle of least privilege is the most important practice to maintain in a Data Lake. This principle ensures that data can only be accessed by those who need access to it. This stops everyone in an organization from having access to all the information in a Data Lake, such as Personally Identifiable Information, or PII.
ZoningMany organizations divide their Data Lake information into different zones, to make granting access and permissions much easier. Organizations will usually form four zones which are called the temporal, raw, trusted, and refined zones. The temporal zone holds temporary data that does not require long term storage. The raw zone holds data that is sensitive and unencrypted, before it has been processed and is secure. The trusted zone holds data that has been deemed secure and is ready to be used in applications. Anyone needing processed data, such as end users, will find it in the trusted zone of the Data Lake. The final zone, the refined zone, holds data that has been run through other applications and returned here as a final output.
Data EncryptionOne important step to securing Data Lakes is the use of data encryption. By following compliance guidelines, such as the Federal Information Processing Standards (FIPS), the most advanced encryption algorithms can be selected for your Data Lake.
SIEM Tool UseSecurity Information and Event Management (SIEM) tools and software work to detect threats, ensure compliance, and manage any other security issues in an organization’s Data Lake. These tools assist companies with providing the highest level of Data Lake Protection possible by finding threats within an IT infrastructure before those threats can compromise data.

EC Data Lakes

A great way to begin protecting your organization’s Data Lake is by utilizing Encryption Consulting’s training sessions. At Encryption Consulting, we offer a variety of training services, including learning to use AWS’ Data Protection Service, GCP’s Key Management Services, and Microsoft Azure’s Key Vault. We can also help install and configure Hardware Security Modules to protect your data encryption keys. Our Cloud Utility Functions, Cloud Data Protector and Bucket Protector, were created specifically with Cloud Data Lake Protection in mind. Cloud Data Protector encrypts on-premises data before it is sent to Google Cloud Platform. Bucket Protector works within Google Cloud Platform itself, to encrypt data as it is uploaded into buckets. Encryption advisory services and enterprise encryption platform implementation services are also offered.

Conclusion

As you can see, Data Lakes provide organizations with a multitude of benefits, from making processes simpler to cutting back on infrastructure costs. By creating zones, using SIEM tools and software, and following the principle of least privilege, your data lake will stay secure from any attempted compromise. To learn more about the services Encryption Consulting can offer you, visit our website: www.encryptionconsulting.com .

Tags:

About the Author

Riley Dickens

Free Downloads

Datasheet of Encryption Consulting Services

Encryption Consulting is a customer focused cybersecurity firm that provides a multitude of services in all aspects of encryption for our clients.

Download

Category you want ...

Cloud Key Management

Cloud-based vs On-premises HSMs

by Dipanshu Bhatnagar

Posted on October 24, 2020Last updated on August 18, 2022

Read Time: 08 min.

Encryption is one of the basic building blocks for any organization containing sensitive data/information. Sensitive data compliant with data privacy regulations creates a brand value for your organization as your organization becomes less prone to data breaches. As we all know the strength of the encryption depends upon two critical factors

Key length
Security of the Keys

Key length is quantifiable and could be determined using the various encryption algorithms such as AES-128 or AES-256. On the other hand, the Security of the key is a subjective matter. As we all know, the more secure the keys are, private keys in asymmetric and shared keys in symmetric encryption, the more powerful the encryption landscape is.

When it comes to the Security of Keys, the best bet is to use HSMs (Hardware Security Module) which are NIST compliant i.e. FIPS-140-2-Level3.

Cloud-based HSM vs. On-Premises HSM

In today’s article, we will compare Cloud-based HSM and On-prem HSM and try to

Find an answer for what criteria a customer should choose as the appropriate option for their organization’s crypto security.

As organizations step up their cloud journey as fast as possible to utilize the advantages of the cloud e.g. scalability, flexibility, cost-effectiveness, they have to parallelly think about data security in their IT landscape. This makes encryption, and subsequently HSMs, an inevitable component of an organization’s Cybersecurity strategy.Based on the use cases, we can classify HSMs into two categories: Cloud-based HSMs and On-Prem HSMsIn regards to the classification of HSMs (On-prem vs Cloud-based HSM), kindly be clear that the cryptographic technology is the same, but delivered via different methods.

On-prem HSMs are specifically useful for storing encryption keys when the organization wants complete control over their keys and policies without having any dependency on the Cloud Service Provider (CSPs). However, this comes at substantial upfront investment in terms of hardware, skilled resources, management software licenses managing the HSM cluster etc.

On-prem HSMs also make sense when an organization uses a secure application which is extremely sensitive to latency. The secure application uses an On-prem HSM only, thus avoiding the latency. Another important use case is where an application with intensive cryptographic operations is in use due to security best practices, technological designs, and/or performance requirements. On-prem HSM is also beneficial to organizations which operate in countries with strict regulatory/compliance requirements on data localization, and where Cloud Service Providers (CSP) may not have a local datacenter in that geographic location. It also benefits organizations with foreseeable workloads, where it is highly unlikely that the business requirements and transaction volumes will exceed the capacity of the HSM in the near future.

On the other hand, Cloud-based HSMs offer out-and-out advantages of the cloud in addition to conventional features of HSMs. To dig deeper, we can further classify the Cloud-based HSM into two categories: Public Cloud HSM Services and Third- Party HSM Services. Some Public Cloud HSM Services offer Single-tenant/dedicated or Multi-tenant services (e.g. AWS, Azure) whereas others offer only Multi-tenant services (e.g. GCP KMS, Oracle Key Vault) thus, these HSM Services are best suited for organizations which are dependent upon single Cloud Service Provider (CSP). In Third Party HSM Services, you can leverage multi-cloud platforms managed through the central management portal (e.g. DPoD) thus, these HSM Services are best suited for organizations with multi-cloud strategies. These HSM Services also offer use-case-based modular services to lessen data protection cost. Some examples of these services are Key Vault, Oracle TDE (Transparent Data Encryption), Code/Digital Signing etc.

Cloud-based HSMs are extremely helpful in the case of SMB (small and medium business) organizations which already have some other IT service dependencies, and substantial upfront investments for On-prem HSMs may not be feasible in terms of cost- effectiveness. Another classic Cloud-based HSM use case is where enterprises want to test or pilot multiple vendor HSM services with minimal upfront investments before committing to a vendor. Also, it is useful in organizations where the workloads are less in the organization/department and application performance and latency requirements are not stringent enough to require a dedicated, On-prem HSM. This model is suitable for smaller organizations which prefer a foreseeable and PAYG (pay-as-you-go) based financial model offered by the Cloud Service Provider (CSP) rather than high initial capital investments required by an On-prem HSM. Organizations/departments with highly variable workloads which might require elasticity (i.e. scaling up and scaling down of the HSM infrastructure) do come under the umbrella of Cloud-based HSMs as well.