3.1 Security Implications of Architecture Models

Responsibility Matrix

shared responsibility between the consumer and the

cloud service provider (CSP)

Cloud

multiple users were given small slices of the computer’s time to run whatever program they needed at that time. Today, a cloud provider typically delivers computing, storage, databases, and an entire assortment of applications online to users who access them from a web browser.

Infrastructure as a service (IaaS)

delivers the computer infrastructure in a hosted service model over the Internet. This method of cloud computing allows the client to outsource everything that would normally be found in a typical IT department

Platform as a service (PaaS)

delivers a platform to develop and manage applications over the Internet, without downloads or installation. PaaS systems are often development platforms designed to operate specifically in the cloud environment, and with them, developers do not need to worry about any servers

Software as a service (SaaS)

delivers a licensed application to customers over the Web for use as a service on demand. The online services are delivered from the Web, and only a web browser is required to interact with them. A SaaS vendor hosts an application and allows a customer to use the application, usually for a set subscription period, after which the application becomes inactive if not renewed

Hybrid Considerations

combination of public and private. This approach allows an organization to leverage the advantages of both environment types. A hybrid cloud environment is the best choice when an organization offers services that need to be configured for diverse vertical markets or has varying needs

Infrastructure as Code (IaC)

enables infrastructure configurations to be incorporated into application code. The way this works is that IaC is code/script in a file, so you can build and destroy entire infrastructures by running the code/script file.

DevOps

set of practices that combines software

development (Dev) and information technology operations

(Ops) with the goal of shortening the system’s development

lifecycle and providing continuous delivery with high

software quality.

function as a service (FaaS)/ serverless

provides on-demand capabilities without the need for you to manage, monitor, or maintain any servers. The cloud provider does that. The purpose of FaaS is to run your application logic when you need it done without you needing to worry about servers

Microservices

offer a newer way to develop applications and services and are really the primary way most applications and services today are developed, especially web-based applications and services

Network Infrastructure

refers to all the devices, hardware, software, connectivity (physical and logical), as well as protocols and services that make up the network

Physical Isolation

pertains to the strategies and designs used to physically separate different components, systems, or environments from each other within a network

Air-gapped

the network segment is restricted completely from access and requires a very specific way to access it

Logical Segmentation

the practice of dividing a network or computing environment into multiple, distinct segments or zones, using software-based solutions instead of physical-based solutions

Software-Defined Networking (SDN)

enables organizations to manage network services through a decoupled infrastructure that allows for quick adjustments to changing business requirements. In the SDN architecture, the control and data planes are decoupled, whereas in a non-SDN architecture they are not decoupled

On-Premises

the technology footprint you maintain internally within your company, hosted by you, and not by a third-party vendor or within a cloud

Centralized architecture

system’s control, operations, and resources are managed from a single, central location

Decentralized architecture

control and resources are distributed across multiple locations or nodes. Each node in the network operates independently and, in some cases, can perform the same functions

Containerization

consists of only the application and its dependencies. Therefore, containerization is a lightweight and efficient form of virtualization that allows for the isolation and packaging of software code, along with all its dependencies, so that it can run uniformly and consistently on any infrastructure

Virtualization

offers attractive cost benefits by decreasing the number of physical machines—both servers and desktops—required in an environment. On the client side, the capability to run multiple operating environments enables a machine to support applications and services for an operating environment other than the primary environment

Hypervisor

software that runs on a bare-metal server or within the operating system of a server that enables you to create VMs and control the resources (CPU, memory, storage, and NIC) that the VMs have access to.

Type 1 hypervisor, or bare-metal hypervisor

software that runs directly on the host’s hardware to control the hardware and to manage guest operating systems (VMs). For this reason, they are considered more efficient and secure than Type 2 hypervisors

Type 2 hypervisor, or hosted hypervisor

software that runs on a conventional operating system, just like other computer programs, to manage VMs. Type 2 hypervisors abstract guest operating systems from the host operating system

IoT ( Internet of Things)

are interconnected physical objects equipped with sensors, software, and other technologies that enable them to collect and exchange data with other devices and systems over the Internet or other communication networks

Industrial control systems (ICSs)

general term for various types of control systems and associated instrumentation for critical systems across a number of sectors, such as infrastructure, facilities, industrial, logistics, and energy

Supervisory control and data acquisition (SCADA)

subset of ICS. Therefore, an ICS is managed via a SCADA system that provides a human– machine interface (HMI) for operators to monitor the status of the system

Real-Time Operating System (RTOS)

small operating system used in embedded systems and IoT applications. The primary purpose of an RTOS is to allow the rapid switching of tasks, with a focus on timing instead of throughput, allowing applications to run with precise timing and high reliability

Embedded Systems

specialized computing systems that perform dedicated functions and/or are designed for specific control applications within a larger system. Unlike general-purpose computers that can run a variety of applications, embedded systems are task-specific, running predefined tasks with very specific requirements

High Availability

refers to an architectural design concept that aims to ensure a 100% level of operational performance and reliability (uptime) during a predefined period for services, applications, or systems. This design concept is crucial for critical systems where downtime has significant consequences, such as in financial services, healthcare, and telecommunications

Availability

refers to the degree by which a system, service, or data is accessible and functional when needed by users. This requires you to implement redundant systems, failover mechanisms, and robust infrastructure designs to minimize downtime and maintain continuous operations

Resilience

the ability to recover from failure, and hopefully doing so without impacting service. There are two types of equipment: equipment that has failed and equipment that has not yet failed, but will fail soon. Therefore, you need to be prepared

Cost

the financial impact of implementing, maintaining, and operating the architecture. Your cost considerations have both a direct and indirect impact on your ability to protect your assets effe

Responsiveness

the ability of a system to react quickly to inputs or changes. Factors affecting responsiveness include network latency, system load, resource optimization, and the efficiency of the codebase

Scalability

the opportunity for expanding (adding resources to handle increased load) and contracting (removing resources when they’re no longer needed) within a system, service, or architecture. Scalability is critical for meeting the needs of users while also maximizing/minimizing resource utilization

Ease of Deployment

refers to the simplicity and speed with which an architecture, system, or service can be deployed or updated

Risk Transference

the act of shifting risk from one party to another, for example, transferring the physical risk to a cloud provider, or transferring the financial impact of a breach to an insurance company

Ease of Recovery

refers to how straightforward it is to restore services and data after a disruption. The ability to recover from any issue or service disruption is a major consideration when planning an architecture, especially one that needs to remain secure

Patch Availability

the accessibility of software updates provided by vendors or developers to address vulnerabilities, bugs, or enhance functionality within their products

Inability to Patch

situations where software or systems cannot be updated with the latest patches or security fixes for various reasons

Power

the electrical energy required to operate computer systems, servers, network devices, and the infrastructure supporting them, including the all-important cooling systems. The management and provisioning of power are critical to ensuring all systems remain operational, secure, and resilient against various threats

Compute

the processing requirements necessary to execute software instructions, manage data, and perform computational tasks for running applications and services