What if you could manage routing, DNS, DHCP, NAT, IPSEC VPN, SSL VPN, deploy IDS/IPS, Firewall the network. Configure forward and reverse proxy, authenticate users via Radius or Mobile OTP, integrate AD / LDAP user accounts, manage SSL certificates and using single platform and dashboard. PFsense brings power of varied networking services under one hood.
fSense as Virtual Appliance
FreeBSD based open source pfSense, is an excellent combination of network and security features. 14 years of continuous development and deployment in production networks, pfSense is now shaped into a swiss army knife of routing, security and other networking services such as DNS, DHCP, packet capturing, VPN services and much more. The ‘much more’ part is coming from the ease of adding other well-known, opensource network tools into pfSense platform as packages. PfSense comes in hardware, software and cloud deployment modes. Current company behind development and maintenance of pfSense is Netgate, which offers hardware and official support. Let’s have look on some of the capabilities of pfsense which can support a fully functional, secure and flexible network.
- Free, cost effective
- CPU, RAM, HDD can be increased as demand increases
- Minimum resources, maximum utilization
- Increase number of interfaces as needed
- Can be deployed on cloud platform
Tables, Primary Keys, and Queries
Cassandra’s data model is a partitioned row store. Rows are organized into tables. Tables have primary keys and columns. It sounds a bit RDBMS-like, but it starts to differ. The first part of a primary key is the partition key. Then, within a partition, rows are clustered by the remaining columns of the primary key.
In Cassandra, queries are optimized by primary key definition. To a large extent, it’s the design of the primary key that supports the WHERE clause of a query. Because Cassandra is by design a distributed database, efficiency is gained when data is grouped tightly together on nodes by partition. This grouping is defined by the primary key.
F5 Lab Setup
This F5 lab consists of 2x virtual appliances, with all modules available for configuration. It also includes 3x small webservers.
Conclusion
Data modelling has always been a cornerstone of databases. Conceptual data modelling and relational database design have been extensively studied and are generally well understood. Unfortunately, the majority of relational data modelling techniques are not transferrable to NoSQL databases.
NoSQL modelling remains a challenging and open problem. But you can start by knowing your use cases and focus your design to support them, including data replication for your most frequently used queries, you’ll be off to a good start.
