The advertising ecosystem is a broad entity made up of many different processes and platforms, and just as many acronyms. One of its most influential additions over the past decade has been the introduction of Real-Time Bidding (RTB).
One of the most influential additions to the advertising ecosystem over the past decade has been demand-side platforms (DSPs).
What Is A Demand-Side Platform (DSP)?
A demand-side platform (DSP) is an advertising technology (AdTech) platform that allows advertisers working at brands and ad agencies to buy inventory (aka ad space) on an impression-by-impression basis from publishers via supply-side platforms (SSPs) and ad exchanges. DSPs enable media buyers (advertisers and agencies) to purchase a range of inventory from many different publishers all from one user interface.
An easy way to understand how DSPs operate is to think of them as stockbrokers. In the same way investors use stockbrokers to purchase stock from companies via the stock exchange, media buyers use DSPs to purchase ad inventory from publishers through the ad exchange.
Demand-side platforms started to emerge in when real-time bidding began. Back then, most of the inventory purchased was for web browsers in desktops and laptops, but DSPs nowadays allows media buyers to purchase inventory programmatically across a range of devices (e.g. smartphone and tablets) and mediums, including mobile, video, and native.
A DSP is essentially a middle man, but they are responsible for far more than just the purchasing of available inventory.
Here are some of the main functions DSPs provide media buyers:
How Do Demand-Side Platforms (DSPs) Work?
Once an advertisers has set up their campaigns in a DSP, including targeting and creatives (ads), the DSP then bids on impressions offered by ad exchanges and SSPs. So each time a DSP receives a request from an ad exchange or SSP telling it that there is an impression available, the DSP analyzes the data about the user and decides how much this particular user is worth based on their relevance to the media buyer.
Let&#;s look at this a bit more through the use of an example:
A user, who is about 30 years old and a big sports fan living in New York, accesses a website. The website sends an ad request to the supply-side platform (SSP), which then sends the request to the ad exchange. The ad exchange then tells the DSPs that an impression is available on a website and starts the bidding process.
Let&#;s take a step back for a moment and imagine the DSPs have the following targeting parameters set up:
While all of the above DSPs could potentially gain something out of displaying their ad to the user, it is DSP 3 that would gain the most from this ad, as the user fits the target audience perfectly.
As a result, the DSPs would evaluate the ad, match it against their data and target parameters, and would bid on the impression based on this information. In this case, the bidding would probably look something like this:
(CPM = cost per mille, the common unit in advertising, which means cost per thousand impressions. The actual cost of the bid is 1/ the cost of the CPM price, as the advertiser is bidding on a single impression, not purchasing a thousand impressions at once.)
Traditionally, ad exchange and SSPs transact using the second-price auction model. In other words, the winner pays the second highest bid price for the impression plus $0.01.
However, over the past few years, the auction dynamics of RTB transactions has changed slightly, with some ad exchanges and SSPs incorporating a first-price auction model.
We&#;ve written about the different auction dynamics here and here.
Once an impression is sold, it is sent back to the website and is displayed to the user. This process occurs each time a user accesses a website or refreshes the page.
It&#;s important to note that this bidding process happens within the ad exchange or SSP in real time, hence the name real-time bidding, and takes roughly 100 milliseconds to complete.
Our AdTech development teams can work with you to design, build, and maintain a custom-built demand-side platform (DSP) for any programmatic advertising channel.
Learn moreWhat Are the Main Components Of a Demand-Side Platform?
While no two DSPs are exactly the same, most will include the same types of components.
Here&#;s a visual representation of the main ones:
We explain what these components do in one of our previous posts.
What Advantages Do DSPs Offer Media Buyers?
DSPs traditionally were used as a way to buy remnant (unsold) inventory from publishers, but they are now becoming a way of purchasing available and even premium inventory.
The main reason for this shift in purchasing methods is due to the numerous advantages DSPs provide media buyers. Here are some of the main benefits:
The rapid uptake of RTB has seen the number of DSPs dramatically increase over the recent years and with the opportunity to build DSPs that incorporate new integrations and advanced features & technology, their popularity is set it increase even further.
How Does Targeting Work In a DSP?
There are a number of ways an advertiser can run targeted ad campaigns with a DSP, but at the heart of it all is data.
For example, an advertiser could target users based on the following data:
Behavioral data: Includes information about the user&#;s behavior and interests, such as which websites they&#;ve visited, what products they&#;ve purchases, which ads they&#;ve interacted with, etc.
Contextual data: Includes information about the website or mobile app, such as URL, categories, and the content on the page.
Demographic data: Includes information about the user&#;s location, age, job title, gender, and so on.
But how does a DSP access these types of information?
Typically, behavioral and demographic data would need to be imported from a third-party data broker or DMP, which would have been collected from a range of online and offline sources. The DSP would sync cookies with the DMP to exchange user data, which can then be used for targeting.
Other types of data, such as contextual data or even user agent data (e.g. browser type, device advertising ID and operating system), are usually passed along during the bid request from the ad exchanges or SSPs.
Who Are the Main DSP Companies?
Over the past decade, dozens of DSP companies have entered the AdTech landscape, with many specializing in specific areas, such as video or native.
Below are some of the most well-know DSP companies:
Sizmek is one of the longest-standing AdTech companies in the industry (originally founded under the name Eyeblaster). Apart from building their own solutions, they&#;ve also acquired a number of other advertising technology companies over the years, including Rocket Fuel for $145 million back in . MediaMath, founded in , is one of the first DSPs to emerge the market, and one that often comes up in the context of online advertising. The company provides advertisers and marketers with tools to buy ads online through a single interface. Their integrated DSP and DMP is exclusively buy-side aligned. The company prides itself on alignment and transparency, multi-channel and full-funnel approach, performance, global scale and extensibility. Integration with MediaMath offers a number of benefits which we have discussed in another post on our blog. AdForm, with its headquarters in Copenhagen, Denmark, and offices in other locations around the globe, provides advertisers and agencies with an open and transparent advertising technology platform. Their DSP allows media buyers to purchase inventory across different formats and improve campaign performance with the help of their real-time algorithmic optimizations. BrightRoll, formerly owned by Yahoo and acquired by Oath in , is an advertising platform which includes a real-time bidding marketplace and powers programmatic video. It is used by brands, agencies, agency trading desks, demand-side platforms (DSPs), and advertising networks &#; and enables them to connect with digital audiences to support advertising campaign objectives. Criteo is a DSP providing personalized retargeting. The platform serves personalized online display advertisements to consumers who have previously visited the advertiser&#;s website. The company currently operates in a total of 30 markets around the world and is headquartered in Paris, France. DoubleClick Bid Manager (aka DBM) is a leading demand-side platform (DSP) from Google. It offers agencies, trading desks, and advertisers access to the world&#;s most exclusive collection of display, video, native and mobile inventory available in real-time.The Benefits of Building a Custom Demand-Side Platform (DSP)
While there are a number of DSPs companies on the market that offer a range of features for advertisers and agencies, there are a number of benefits in developing your own custom DSP:
Ownership of data and technology: By building a custom DSP, you can gain control of both the technology and data. This is particularly beneficial for advertisers with large advertising campaigns and agencies that purchase large amounts of inventory on behalf of their clients.
Elimination of white-label fees and commissions: While building your own custom DSP won&#;t eliminate all of the fees and commissions involved in the online media transactions, it will certainly eliminate the known fees and commissions associated with using white-labeled solutions, which can be quite a costly expense for a lot of media buyers.
Control of the product&#;s roadmap: Most DSPs come packed with features, but they are often not a one-size-fits-all platform. For brands and agencies with specific use cases, developing a DSP enables them to have full control over the product&#;s roadmap and build the features that they want and need.
Our AdTech development teams can work with you to design, build, and maintain a custom-built demand-side platform (DSP) for any programmatic advertising channel.
Learn moreA digital signal processor (DSP) is a specialized microprocessor chip, with its architecture optimized for the operational needs of digital signal processing.[1]:&#;104&#;107&#;[2] DSPs are fabricated on metal&#;oxide&#;semiconductor (MOS) integrated circuit chips.[3][4] They are widely used in audio signal processing, telecommunications, digital image processing, radar, sonar and speech recognition systems, and in common consumer electronic devices such as mobile phones, disk drives and high-definition television (HDTV) products.[3]
The goal of a DSP is usually to measure, filter or compress continuous real-world analog signals. Most general-purpose microprocessors can also execute digital signal processing algorithms successfully, but may not be able to keep up with such processing continuously in real-time. Also, dedicated DSPs usually have better power efficiency, thus they are more suitable in portable devices such as mobile phones because of power consumption constraints.[5] DSPs often use special memory architectures that are able to fetch multiple data or instructions at the same time.
[
edit
]
A typical digital processing systemDigital signal processing (DSP) algorithms typically require a large number of mathematical operations to be performed quickly and repeatedly on a series of data samples. Signals (perhaps from audio or video sensors) are constantly converted from analog to digital, manipulated digitally, and then converted back to analog form. Many DSP applications have constraints on latency; that is, for the system to work, the DSP operation must be completed within some fixed time, and deferred (or batch) processing is not viable.
Most general-purpose microprocessors and operating systems can execute DSP algorithms successfully, but are not suitable for use in portable devices such as mobile phones and PDAs because of power efficiency constraints.[5] A specialized DSP, however, will tend to provide a lower-cost solution, with better performance, lower latency, and no requirements for specialised cooling or large batteries.[citation needed]
Such performance improvements have led to the introduction of digital signal processing in commercial communications satellites where hundreds or even thousands of analog filters, switches, frequency converters and so on are required to receive and process the uplinked signals and ready them for downlinking, and can be replaced with specialised DSPs with significant benefits to the satellites' weight, power consumption, complexity/cost of construction, reliability and flexibility of operation. For example, the SES-12 and SES-14 satellites from operator SES launched in , were both built by Airbus Defence and Space with 25% of capacity using DSP.[6]
The architecture of a DSP is optimized specifically for digital signal processing. Most also support some of the features of an applications processor or microcontroller, since signal processing is rarely the only task of a system. Some useful features for optimizing DSP algorithms are outlined below.
[
edit
]
[
edit
]
By the standards of general-purpose processors, DSP instruction sets are often highly irregular; while traditional instruction sets are made up of more general instructions that allow them to perform a wider variety of operations, instruction sets optimized for digital signal processing contain instructions for common mathematical operations that occur frequently in DSP calculations. Both traditional and DSP-optimized instruction sets are able to compute any arbitrary operation but an operation that might require multiple ARM or x86 instructions to compute might require only one instruction in a DSP optimized instruction set.
One implication for software architecture is that hand-optimized assembly-code routines (assembly programs) are commonly packaged into libraries for re-use, instead of relying on advanced compiler technologies to handle essential algorithms. Even with modern compiler optimizations hand-optimized assembly code is more efficient and many common algorithms involved in DSP calculations are hand-written in order to take full advantage of the architectural optimizations.
For more information, please visit interwiser.
[
edit
]
[
edit
]
[
edit
]
[
edit
]
[
edit
]
DSPs are usually optimized for streaming data and use special memory architectures that are able to fetch multiple data or instructions at the same time, such as the Harvard architecture or Modified von Neumann architecture, which use separate program and data memories (sometimes even concurrent access on multiple data buses).
DSPs can sometimes rely on supporting code to know about cache hierarchies and the associated delays. This is a tradeoff that allows for better performance[clarification needed]. In addition, extensive use of DMA is employed.
[
edit
]
DSPs frequently use multi-tasking operating systems, but have no support for virtual memory or memory protection. Operating systems that use virtual memory require more time for context switching among processes, which increases latency.
[
edit
]
TRW TDC multiplier-accumulator[
edit
]
In , Richard Wiggins proposed the Speak & Spell concept to Paul Breedlove, Larry Brantingham, and Gene Frantz at Texas Instruments' Dallas research facility. Two years later in , they produced the first Speak & Spell, with the technological centerpiece being the TMS,[15] the industry's first digital signal processor. It also set other milestones, being the first chip to use linear predictive coding to perform speech synthesis.[16] The chip was made possible with a 7 μm PMOS fabrication process.[17]
In , American Microsystems (AMI) released the S.[3][4] The AMI S "signal processing peripheral", like many later DSPs, has a hardware multiplier that enables it to do multiply&#;accumulate operation in a single instruction.[18] The S was the first integrated circuit chip specifically designed as a DSP, and fabricated using vertical metal oxide semiconductor (VMOS, V-groove MOS), a technology that had previously not been mass-produced.[4] It was designed as a microprocessor peripheral, for the Motorola ,[3] and it had to be initialized by the host. The S was not successful in the market.
In , Intel released the as an "analog signal processor".[19] It had an on-chip ADC/DAC with an internal signal processor, but it didn't have a hardware multiplier and was not successful in the market.
In , the first stand-alone, complete DSPs &#; Nippon Electric Corporation's NEC μPD based on the modified Harvard architecture[20] and AT&T's DSP1 &#; were presented at the International Solid-State Circuits Conference '80. Both processors were inspired by the research in public switched network (PSTN) telecommunications. The μPD, introduced for voiceband applications, was one of the most commercially successful early DSPs.[3]
The Altamira DX-1 was another early DSP, utilizing quad integer pipelines with delayed branches and branch prediction.[citation needed]
Another DSP produced by Texas Instruments (TI), the TMS presented in , proved to be an even bigger success. It was based on the Harvard architecture, and so had separate instruction and data memory. It already had a special instruction set, with instructions like load-and-accumulate or multiply-and-accumulate. It could work on 16-bit numbers and needed 390 ns for a multiply&#;add operation. TI is now the market leader in general-purpose DSPs.
About five years later, the second generation of DSPs began to spread. They had 3 memories for storing two operands simultaneously and included hardware to accelerate tight loops; they also had an addressing unit capable of loop-addressing. Some of them operated on 24-bit variables and a typical model only required about 21 ns for a MAC. Members of this generation were for example the AT&T DSP16A or the Motorola .
The main improvement in the third generation was the appearance of application-specific units and instructions in the data path, or sometimes as coprocessors. These units allowed direct hardware acceleration of very specific but complex mathematical problems, like the Fourier-transform or matrix operations. Some chips, like the Motorola MC, even included more than one processor core to work in parallel. Other DSPs from are the TI TMS320C541 or the TMS 320C80.
The fourth generation is best characterized by the changes in the instruction set and the instruction encoding/decoding. SIMD extensions were added, and VLIW and the superscalar architecture appeared. As always, the clock-speeds have increased; a 3 ns MAC now became possible.
[
edit
]
Modern signal processors yield greater performance; this is due in part to both technological and architectural advancements like lower design rules, fast-access two-level cache, (E)DMA circuitry, and a wider bus system. Not all DSPs provide the same speed and many kinds of signal processors exist, each one of them being better suited for a specific task, ranging in price from about US$1.50 to US$300.
Texas Instruments produces the C series DSPs, which have clock speeds of 1.2 GHz and implement separate instruction and data caches. They also have an 8 MiB 2nd level cache and 64 EDMA channels. The top models are capable of as many as MIPS (millions of instructions per second), use VLIW (very long instruction word), perform eight operations per clock-cycle and are compatible with a broad range of external peripherals and various buses (PCI/serial/etc). TMS320C chips each have three such DSPs, and the newest generation C chips support floating point as well as fixed point processing.
Freescale produces a multi-core DSP family, the MSC81xx. The MSC81xx is based on StarCore Architecture processors and the latest MSC DSP combines four programmable SC StarCore DSP cores. Each SC StarCore DSP core has a clock speed of 1 GHz.
XMOS produces a multi-core multi-threaded line of processor well suited to DSP operations, They come in various speeds ranging from 400 to MIPS. The processors have a multi-threaded architecture that allows up to 8 real-time threads per core, meaning that a 4 core device would support up to 32 real time threads. Threads communicate between each other with buffered channels that are capable of up to 80 Mbit/s. The devices are easily programmable in C and aim at bridging the gap between conventional micro-controllers and FPGAs
CEVA, Inc. produces and licenses three distinct families of DSPs. Perhaps the best known and most widely deployed is the CEVA-TeakLite DSP family, a classic memory-based architecture, with 16-bit or 32-bit word-widths and single or dual MACs. The CEVA-X DSP family offers a combination of VLIW and SIMD architectures, with different members of the family offering dual or quad 16-bit MACs. The CEVA-XC DSP family targets Software-defined Radio (SDR) modem designs and leverages a unique combination of VLIW and Vector architectures with 32 16-bit MACs.
Analog Devices produce the SHARC-based DSP and range in performance from 66 MHz/198 MFLOPS (million floating-point operations per second) to 400 MHz/ MFLOPS. Some models support multiple multipliers and ALUs, SIMD instructions and audio processing-specific components and peripherals. The Blackfin family of embedded digital signal processors combine the features of a DSP with those of a general use processor. As a result, these processors can run simple operating systems like μCLinux, velocity and Nucleus RTOS while operating on real-time data. The SHARC-based ADSP-210xx provides both delayed branches and non-delayed branches.[21]
NXP Semiconductors produce DSPs based on TriMedia VLIW technology, optimized for audio and video processing. In some products the DSP core is hidden as a fixed-function block into a SoC, but NXP also provides a range of flexible single core media processors. The TriMedia media processors support both fixed-point arithmetic as well as floating-point arithmetic, and have specific instructions to deal with complex filters and entropy coding.
CSR produces the Quatro family of SoCs that contain one or more custom Imaging DSPs optimized for processing document image data for scanner and copier applications.
Microchip Technology produces the PIC24 based dsPIC line of DSPs. Introduced in , the dsPIC is designed for applications needing a true DSP as well as a true microcontroller, such as motor control and in power supplies. The dsPIC runs at up to 40MIPS, and has support for 16 bit fixed point MAC, bit reverse and modulo addressing, as well as DMA.
Most DSPs use fixed-point arithmetic, because in real world signal processing the additional range provided by floating point is not needed, and there is a large speed benefit and cost benefit due to reduced hardware complexity. Floating point DSPs may be invaluable in applications where a wide dynamic range is required. Product developers might also use floating point DSPs to reduce the cost and complexity of software development in exchange for more expensive hardware, since it is generally easier to implement algorithms in floating point.
Generally, DSPs are dedicated integrated circuits; however DSP functionality can also be produced by using field-programmable gate array chips (FPGAs).
Embedded general-purpose RISC processors are becoming increasingly DSP like in functionality. For example, the OMAP3 processors include an ARM Cortex-A8 and C DSP.
In Communications a new breed of DSPs offering the fusion of both DSP functions and H/W acceleration function is making its way into the mainstream. Such Modem processors include ASOCS ModemX and CEVA's XC.
In May , Huarui-2 designed by Nanjing Research Institute of Electronics Technology of China Electronics Technology Group passed acceptance. With a processing speed of 0.4 TFLOPS, the chip can achieve better performance than current mainstream DSP chips.[22] The design team has begun to create Huarui-3, which has a processing speed in TFLOPS level and a support for artificial intelligence.[23]
[
edit
]
[
edit
]
If you want to learn more, please visit our website Digital Signal Processing DSP.