Good morning and good evening, ladies and gentlemen. Thank you and welcome to VNET Group Inc, Fourth Quarter, 2021 Earnings Conference Call. At this time, all participants are in listen-only mode. We will be hosting a question-and-answer sessions after the management prepare remarks. With us today are Mr. Samuel Shen, Chief Executive Officer and Executive Chairman of Retail IDC. Mr. Tim Chen, Chief Financial Officer, and Ms. Xinyuan Liu, Investor Relations Director of the company. I will now turn the call over to the first speaker today, Ms. Liu, IR Director of VNET Group Inc. Please go ahead, ma'am.

Hello, everyone. Welcome to our Fourth Quarter 2021 Earnings Conference Call. Our earnings release was distributed earlier today and you can find a copy on our IR website. Please note that the discussion today will contain forward-looking statements made under the Safe Harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements are subject to risks and uncertainties that may cause actual results to differ materially from our current expectations. For detailed discussions of these risks and uncertainties, please refer to our latest Annual Report and other documents filed with the SEC. VNET does not undertake any obligations to update any forward-looking statements, except as required under applicable laws. As a reminder, this conference is being recorded. In addition, a webcast of this conference call will also be available on our IR website at ir.vnet.com. I will now turn the call over to our CEO, Samuel.

Thank you, Xinyuan. Good morning, and good evening, everyone. Thank you for joining our fourth quarter 2021 earnings conference call. We concluded 2021 with strong operating and financial results. Operationally, we successfully achieved this year's delivery target by adding approximately 25,000 cabinets in the fourth year of 2021. Including 13,276 cabinets that were delivered in the fourth quarter. Financially, for the full-year of 2021, we grew our revenue by 28% and our adjusted EBITDA by 32%. We attributed our achievements to favorable government policies, robust market demand, persistent strategy execution, and methodical service expansion. First, we're pleased to see the favorable government policies continue to provide a strong tailwind to our industry development. Last month, the eastern data, western computing plan was joined and released by China 's National Development and Reform Commission, together with three other central regulatory departments. Of the eight national computing UPS, we have already deployed our data centers in Beijing, Cangene, Hoby region, Yangtze River Delta, Greater Bay Area, Chengdu-Chongqing economics circle and Inner Mongolia autonomous region. In addition, this January, the state council of China unveil the first five-year gross plan for the digital economy, highlighting the sectors role in reshaping global economic structure and rolling out developed targets through 2025, the plan laid out measures for operating national infrastructure, bolstering the role of data as a production element, and promoting digital transformation. The plan also gives priority to the development of digital infrastructure a pillar of achieving digital economic prosperity that will also spur investment in drive overall economic growth. During the fourth quarter, we received orders not only from fair weather technology companies in the internet sectors, but also traditional companies in brick-and-mortar industries that are transforming through digitalization. ln addition, we're seeing growing demand in both the wholesale and retail segments.…

Thank you, Samuel. Good morning and good evening, everyone. Before we start our detailed financial discussion, please note that we will present non-GAAP measures today. Our non-GAAP results exclude certain non-cash expenses which are not part of our core operations. The details of these expenses may be found in the reconciliation tables included in our press release. Please also note that unless otherwise stated, all the financial numbers we present today are for the fourth quarter of 2021 and in Renminbi terms. While percentage changes are on a year-over-year basis. As Samuel mentioned, our performance in 2021 was characterized by healthy revenue growth and margin expansion, improved operational efficiency, augmented utilization rates and consistent cabinet capacity deliveries which proceeded according to schedule despite macro uncertainties. Net revenue in the fourth quarter of 2021 increased to RMB1.75 billion, a 29.4% year-over-year increase from the fourth quarter of 2020. This increase was mainly due to increased customer demand for our highly scalable carrier and cloud neutral IDC solutions from both wholesale and retail IDC customers, as well as the notable growth in our cloud basis. Gross profit in the fourth quarter of 2021 was $380 million representing a year-over-year increase of 29.1% and a sequential increase of 1.3%. Gross margin in the fourth quarter of 2021 was 21.8% compared to 21.8% in the fourth quarter of 2020, and 24% in the third quarter of 2021. The year-over-year decrease in gross margin or small decrease in gross margin was primarily attributable to the massive delivery of new cabinets, which usually have a ramp up phase to reach the expected profit level. Adjusted cash -- gross profit, which excludes depreciation, amortization and share-based compensation expenses, was $713.8 million in the fourth quarter of 2021, an increase of 22.7% from the fourth quarter of 2020…

Thank you. We will now begin the question-and-answer sessions. [Operator Instructions]. Next question comes from the line of Yang Liu from Morgan Stanley. Please go ahead.

Thank for the opportunity to ask a question, two questions from my side the first one is on that the amount. We see downward revision of those three year new capacity delivery plan for 2022. Could you please update us in terms of the wholesale and retail fleet? And if I remember correctly, previously, 25,000 actually is 80% for wholesale and the scale ratio and 20% for traditional retail business. But once you, the news fleet and the where do we see the downward revision of this amount. The second is for 2022 margin. We see the new, essentially to guidance for revenue growth rate in pretty strong but EBITDA undergrowth revenue. What is the reason for that? Is this mainly because of the higher utility costs, or if there's any other reason behind that? Thank you.

Do you want to take the question first and then I can chime in later?

Absolutely. In terms of the demand outlook, we are looking at a roughly 60/40 split, wholesale versus retail. And so the retail does comprise of component of the scale retail. Actually the downward guide on the cabinets was actually a pro - rata guide, so we're not necessarily saying that there's one segment versus the other segment. And then I guess in terms of the question on the EBITDA versus the revenue growth, we are, as you know, steadily ramping up the wholesale side of the engine, but we also have other business segments. And so some of the other business segments are not EBITDA types of businesses, or they have lower margins, and so that actually is increasing the sort of cost related to those. More importantly, as we also, in our guidance, have factored in the increase in utility cost. And that accounts for, let's say, around just under 1% of the decline in margins as well. So I think that's the main drivers, Yang, to your question. I guess in terms of the overall demand outlook, in terms of some of the details, Samuel, I don't know if you wanted to add some color for Yang on the wholesale and retail demand that we've been seeing.

Again, thanks Yang for the questions. I think overall, given the favorable government policy, and also the rise of data sovereignty and plus the additional transformation momentum. So overall, we're still very bullish about our loan range of outlook in general. But we also have to understand that there is a macro issues that we have to be careful about industry-wise and market-wise. There's also kind of different climate compared to about a last year, and therefore, company-wise, we want to be only cautious. On one hand, there's a decent priority. On the other hand, there's a resource needs that we want to balance, so that's the reason. Overall, business outlook, fantastic, and we're very bullish, but we're taking a cautious approach step-by-step. So I think that's the additional color which I want to add on top of what Tim mentioned earlier. Thank you, Yang.

One more point on the question you had on EBITDA l think what are the other factors that you want to take consideration as well as a fact that we added 13,000 cabinets at the end of 2021 and so we are also factoring drag on the EBITDA when it will take for these cabinets to still the ramp up during the course of 2022. So there will be more costs, obviously in the first part of the year until they start ramping up and these are fixed costs with the deliver cabinets.

Thank you, Matt, follow-up in terms of other low-margin business. Is it the culture for Microsoft and also the VPN growing faster or and also other IT service business or something else has some dragging effect on the margin?

I can probably comment on that one. So first of all, again, precisely, it is not a low margin business. I would say it's not like pure IDC wholesale play, which is EBITDA type of business. Core business, or very added services, they're not much relate to EBITDA, but it is a very healthy business. Again, from the gross margin point-of-view. But assuming you also interest about to know about a Microsoft Cloud gross and also VPN gross. I think from a market perspective, the Microsoft Cloud, which basically includes the Azure, all Phase three to five dynamics VC -25, and part of platform, is highly welcome in the enterprise space. And then of course we're benefiting from their gross by being the Microsoft closest operating partners in China. From number perspective, I would say we continue enjoying the very healthy double-digit growth year-over-year. And then switched to the VPN that you mentioned about, which is POS would V PN. It is still highly regarded as one of the highest quality options for enterprise connection. But we're seeing some of the customers having a needs to balance their price performance. Some of them will continue to go for MTL should VPN, and then the some of them with we try to go with the FDA [Indiscernible] lucky because we do provided both routing options to, for customer to issues. We're seeing the SD wind business because very, very fast, high double-digits growth year-over-year. So I think combined together, it is some still very healthy and great business for us. Hopefully that address your question, Yang

Essential center.

[Operator Instructions] Next question comes from the line of Edison Lee of Jefferies. Please go ahead.

Hi. Thank you very much for the presentation. I want to focus on your target for 2022. You gave a range of 14,400 to 17,400 cabinets. I want to know how comparable this number is versus the previous guidance of 25,000 cabinets. Because obviously, wholesale build-out is not based on cabinets, but based on market want. So could you maybe help us understand how we should actually look at this new guidance range versus the previous 25,000? And number two, is that the moment do you have any thoughts about 2023, because previously it was three-year target. And how should investors think about the 2023 situation based on your understanding of demand right now?

Let me take a crack at this. Edison, thanks for the questions. In terms of the comparability between the cabinets provided, I would say it's on a similar basis so we've not changed the basis in terms of disclosing the number of cabinets. You're absolutely correct that the projects which are wholesale based these are higher density cabinets and so naturally there are less cabinets. Now, in terms of the 2023 and looking beyond l would sort of go back to end of 2020 when we provided the three-year forward look. In terms of cabinets, it was at the completion of what was a extremely successful year for the industry and for us and our peers in 2020. And clearly with a lot of the regulatory headwinds that we saw in 2021 that changed dramatically and has not fully turned around yet here as we start 2022. So I would say an echo Samuel's earlier comments, we are extremely bullish on the sector, we're extremely bullish on the customers requirements. And if you look at it, there is no reduction in terms of the end-users like ourselves using data, storing data, analyzing data, and so the demand is there. But again, we've taken a more cautious approach for 2022. Just given the large number of uncertainties in the macro economic environment. We also look to the market and do see slower than expected move in rates. And so for '22 and we've guided down, all of these projects to a certain extent, some of these projects could be accelerated, so moved from a '23 timetable and moved up into '22. And similarly, things, if we see the pickup during the course of this year, then we would start the '23 projects. And therefore, that would lead to a higher number for '23. But frankly, at March of '22, based on the outlook, I would say that '23 should be higher than '22, but whether we're going back to 25,000 again, I would take 25,000 as a end of 2020 what the market condition was at that time and we're taking a more cautious approach, given the current market conditions and probably don't want to push out a number out this early in the year for something that there is enough visibility on. But again, we can turn on and off the CapEx and therefore, we do expect that we can be quite nimble with regards to '23. Thank you.

Thank you Tim. Can l follow-up on two other issues. Number one, is that, and you disclosed that 15 megawatts of new wholesale projects that you signed in 1Q this year and the 11 megawatt you signed in quarter last year. Is it true that these are all from existing wholesale customers and you did not sign up any brand-new customers in 4Q and 1Q?

Samuel, you want to take the first crack at that?

Yeah. Again. Thank you, Edison for the questions. For the 4Q, from a wholesale perspective, we share about seven megawatts. That was indeed from the existing customers. But something we haven't really said out loud is our scale retail. We also have big wins from scale retail customers in Q4, that was actually Four megawatts, in addition to their previous kind of allocation in contract. For the YQ, because some of the orders coming from existing customer. But we do have the orders coming from our new customers, public cloud providers, and also the e-commerce platform companies, and also states on dedicate cloud and so forth. Again, we still have a few more, I would say, a few more discussion going on. It's hopefully probably in main time frame, where we talk about our Q1 annual release, we could share additional color for our Q1 performance.

Okay. Thank you, Samuel. So the state-owned company, state-owned enterprise cloud service providers, brand new customer; is there right?

Yes.

Okay. Thank you. Sorry. Just one follow-up for team on the power costs. You said that once there will be 1% point impact from my understanding is that order wholesale contracts for all the wholesale contracts you do not include power costs. So that's one of the surpass through. So is that 1% point impact coming from retail customers because you have a 1 to 3-year contracts. So that's why you have a time length in terms of passing on power costs.

So as of the end of 21 Edison, the large majority of our billable capacity was actually bundled with power, and a big portion of that is yes, the wholesale business was still ramping up, so it's a very small percentage. We do expect that by the end of this year, there's sort of unbundled or pass-through proportion will probably increase to, let's say, circle around 30%. And it is the ramp-up of the wholesale which is one portion, but the second one is the extra power costs are being included in the quotation of the renewals, as well as the new retail contracts that we're signing. I think the answer to actually yes to both. It is a rollover of new contracts, but also a ramp up of the wholesale.

For power cost, are you assuming a 20% increase year-on-year on just on the unit power cost? Okay, I'm not talking about total power cost.

Overall, depending on the region, we're expecting between 10% to 20%, so full-year 2021 compared to 2022. And again, that's sort of impact that we're expecting is under 1% to our margins.

Okay. Thank you very much. Yeah, that's it for me.

Thank you.

Thank you for the question. Our next question comes from the line of Cong Sherry off CICC. Please go ahead.

Good morning, management. My question is about the definitive plan and you mentioned that 2022 CapEx plans between RMB 4.5 billion. So how much proportion expected to be using M&A opportunities. And do you see the price of IDC projects in primary market? Show it clean trend this year compared to 2020 and 2021. Thank you.

Okay. Thank you very much Charlotte. Let me take this question and Samuel you can add if there's anything else. For the question in terms of the percentage of M&A, I would look back at '21 and say that roughly 30% of 2021 related to acquisitions of land power, acquisitions of surface companies, and acquisitions of actual data center assets. So as I then look at what we've been planning out for 2022 and providing the four to five billion guidance. We're expecting probably in a range of 15% to 30%. It is a wider range, but that's because some of the M and a projects tend to be larger in size. And so a swing of doing it or not doing it could actually result in a larger percentage change. Now, in Institute question about where do we see pricing of ITC projects. I would say we're still using a guideposts in terms of what we see in a market of around 10 times EBITDA pricing. Obviously, when we do evaluate the projects, it's not purely based on the price, but also on the locations in terms of proximity to our existing data centers, as well as where the customer demands are, in making that evaluation. The other point I would make is that, obviously, the best projects that we were -- the vast majority of our projects are self-built projects. So you'll see a lot more greenfield -- brownfield that we've done in the past to build those into our self-build projects, rather than acquiring EBITDA. So l think on opportunistic basis, they will be some acquisitions of operating data centers. But again, we feel that it is the better use of our capital to not just focus on buying EBITDA, but rather focusing on both, again, where our customers requirements are and making sure that we can put together our product that is very attractive to our customers as well. Thank you.

Thank you for the questions. Our next question will come from Sarah Wang from UBS. Please ask your question.

I have two questions. First, is in regarding a client demand. So ended the government's new, you see the rest competing initiatives, do we see more intentions from our clients on moving into the back on the matter is retail or wholesale clients? And second question is that, I think we have mentioned that 2023 still target should be higher than 2022. Might ask, what are the key assumptions here? What do you think is the key moving factor? So meaning for the 2022 sales target, obviously, you mean either macro, or Internet regulation, or supply chain disruptions. So what other key factor you think or moving our sales target? Thank you.

I want to take the first -- take the question first, and then welcome team to provide additional cutter associated with stat. I think both team and I we mentioned about first of all, what we look at the whole market we are truly bullish and we're bullish for reasons. Number 1, the government is providing a lot of a greater favorable policies. Even further define, provide your direction of guidance for Eastern data, Western computing. On the other hand, government automation about the data is going to be key element for future production. And so the data sovereignty, all of a sudden becomes higher order bids for every single enterprise to consider about for their cloud optimization. And number 3, because I think the COVID basically accelerated diesel transformation. And so nowadays, a lot of our enterprises for them to accelerate their diesel transformation is to go through either CapEx to OpEx, reducing the CapEx and focus on the OpEx in order to resonate with their business gross. On the other hand, they are continuing driving the cloud optimization. But in the past that probably one single positive comp, but the hybrid multi-cloud is going to become a new norm. And that the third one is to tap every single thing as-a-service. And then because the VNET were well-positioned, that we have a dual-core growers to engine industry-leading ones. and so for that one, we're very bullish about the future, the way we see that every single day would see a great momentum. But on the other hand we also have to be very cautious because the macro-environments beyond our control, we're talking about the China use and frenemies situation is a new Nod industry-wise. There's going to be a power quota power tariff and also market-wise customers, there are paying more attention on the cost and there is also possibly a premium that we have paid from a merchant acquisition point-of-view, short period of time and so why we're bullish about the long term, but we're cautious in every single stab would take it. So that's a -- I would say the over overall perspective, and then probably Tim can chime in with some of the additional input that we're seeing every single date and cautious steps that were taken. Tim, do you want to add additional commends?

Absolutely, and I would add to Samuel a dish of dig one layer deeper in terms of both, what would you provided to the market in terms of guidance on cabinets. What's Samuel is talking about in terms of the sort of sales and over and then further to your questions around kind of how things look like in terms of the utilization rate assumptions. And with the delivery that we've made at the end of last year. Obviously, we do expect there to be a dip in terms of utilization rates, both due to large number of deliveries, but also each year we then reset the what cabinets are included under maturity to centers, ramp up and so and so forth. But the desire is also to make sure that we are finding a healthy balance between adding capacity, meeting our customer’s requirements. But also looking at the now and today and seeing where the customers are in terms of ramping up and also where their demands are. And so I think there is the desire to find that balance. And so I think with the reduction in the total number of cabinets, we're also still targeting utilization rates of around 60%. And obviously, to the extent that we see customers ramping up faster than we would speed up the deliveries. And in this case right now, we are taking a more cautious approach because 2022 is not, as I said earlier on, one of the other questions, it's not 2020. And so I think that we do need to find that balance then of making sure we deliver the right number of cabinets to match up with the ramp up and the demands of our end customers.

Thank you for the questions. Next questions comes from the line of Clive Cheung of Credit Suisse. Please go ahead.

Hi. Good morning, management. Thank you for taking my question. Congratulations on the results. My first question is on some -- on margin. We mentioned for certain existing customers, there's increasing trend of taking up more value added services. So my question is, if assuming this is a long-term trend, how is this going to impact our margin, if any? Thank you.

Okay.

Go ahead, Samuel. I'll let you start and then I'll add to that.

Sure. Okay. Again, thank you, Clive, for the question. Yes, we do see a big trend for customers to take on the value-added services. As I said, from the high-level perspective, some of the uniqueness for China compared to the rest of the world, as I said it before, is while the rest of the world -- when enterprise is talking about the cautious formation, they're basically three big players, which is AWS, Microsoft and Google. But in China, we have more than a dozen cloud service providers, so that's one thing, the second thing is why the data, I will say privacy, transparency, computing, date of sovereignty becomes the higher order bids for enterprise to think about. Then they will probably develop a new strategy, other than one single public cloud, they might choose to go with a Hybrid markert cloud in a way that it will making sure that data will be stored at their discretion, so that's the second thing. The third one, because the COVID-19 really impact the whole industry so most of the companies, at least in China, they were so willing to go with the OpEx driven approach instead of the CapEx driven approach. Therefore, other than finding a partner to go with co-location support, they want to have the partner, especially mutual partners. They don't pick side, but providing the post-tax services all the way from co-locations support, the networking capabilities, and plans the networking securities, and also the payer model, because they want to have a single-tenant. They also want to have the O&M. They even want to have the hybrid cloud -- multi-cloud management services to support and sold so force. And so that give VNET to grab opportunity. And we saw that, and we grabbed that as well. The question is whether those very other services are like data center at wholesale play, which is pure EBITDA achievement. And my answer to that is probably not, because those are the software capabilities and service capabilities may have not have enough or a lot of the amortization depreciation of elements. And so, that's a reason you may not be a pure Apple -to - Apple comparison for VNET versus our, I would say IDC peer companies. But luckily, as you probably have heard from some of our peer companies, they are seeing that huge retail tailwinds. They want to switch their focus to the retail engine. And so I think to certain degree, that specific comments, or statements endorsed our growth strategy. So I would say we're happy to have a strategy for the dual-core growth engine, and we're very proud to continue to execute on that strategy. Tim, do you have any additional input you want to make?

No, I think Samuel you gave a good overview of the types of businesses and the value-added services and how it works. So again, I think we can dive into some of the modeling side on the call back then.

Yes. Thank you. And my second question is in terms of the security of resources for our 2022 kind of expansion terms, how much of those are now secured and they actually just yet? Thank you.

So now, what may we've obviously given out the list of the actual pipeline projects and those are secure. We have included a range this year and that's because in terms of an overall timing, I think people were asking about overall timing. We expect that a large portion of these will be actually at the end of the year, and so we've included from our experience already, the possibility that projects either get shifted into '23 or pulled up. And so we've included some others, and those are ones where I think there's uncertainty around the overall timetable from the customer point-of-view. So overall, yes, we have it secured. It really is a matter of timing in terms of whether it comes in this side of '22 or the other side of 2023. Hope that answers questions.

Yes, thank you.

Thank you for the questions. Next question will comes from Ethan Zhang from Nomura Securities, please go ahead.

I note that you mentioned that we'd see that customers seem more value-added service for our retail IDC business. So l just wonder what's the outlook for our retail MRR on the FY22 as going forward. And second question is regarding the new IDC projects. It notes that we have -- just have -- we have some wholesale projects under construction in terms of electrons economic circle. So just want to see your view on rating. I can see demand in that region because I think this was done on new market gratitude. Thanks.

And maybe Samuel, I'll take the MRR question and then you went to take the second question.

Sure.

You're still look in terms of MRR and then I'm happy you've sort of asked what the view of LR4 for 22 is. It will move around quarter-to-quarter as you've seen last year as well. But we do expect that the MRR to stay in 9 thousand plus range, and then trending upwards as there are more and more value-added services that we sell to our customers. Obviously, some of the volatility comes around the fact that not all the customers when they first sign on, is taking on the full package of services. And so that's something which the more cabinets we add, if they're more for what I'll call basic service or basic plus one versus a bundled solution of multiple service., I think that's going to affect the MRR, but I think overall our MRR trend, again, we expect to be in a very positive trend going forward? Let me pass to you Samuel, interpret the questions around the [Indiscernible] region.

So thank you, Ethan, for the questions. I think earlier this year Kevin has new at directional guidance, in terms of Eastern data and also the western computing. So if you look at the eastern data, western computing in Chengdu and Chongqing area has been identified. One of the eight regions, and then that specific region, together with Beijing, Shanghai, Guangzhou, and Shenzhen, we'll see the Tier one city. The four of them are actually defined as to Eastern data. So even though a Chongqing are so like I would say a little bit kind of West compared to Beijing, Shanghai, Shenzhen from Geo perspective, but that specific area does have a strong need from a business and also market perspective. A lot of the scenarios around, I would say low-latency, high bandwidth and also no packet loss. So ideally, for a region specialized to provide the Eastern data per se. So that's also one of the areas that we're going to continue to invest, as a matter of fact, one of the projects in our 2022 is to talk about that specific areas and we're seeing, I would say, great momentum picked up. If I look back in 2021 versus 2020, that's also one of the area start with small but growing fast. So we're very optimistic about the Chengdu and Chongqing areas. So hopefully that address your question, Ethan.

Thank you for the question. Next question comes from Albert Hung of JP Morgan. Please go ahead.

Thank you for taking my question. I want to ask, how should we think about the competitive landscape going forward? Because I see two major factor. One, a common new estate, which is in our west computing is driving the customer into the remote areas, while VNET has vertical, each in Tier one seated. Second, one of your competitors forms to focus more on retail business. So I'm wondering how this two factor will affect your competitive landscape in the future. My second question is, could you comment on the latest pricing in wholesale customer as many of net large internet service provider out doing cost reduction. And you also coming down a bit the future demand outlook. How should we read it as the implication for pricing? Thank you.

Let me start. Thanks for the questions. By the way Albert, let me start with the question around sort of the retail side of the business in terms of, I guess you've mentioned one of our peers wants to focus on the retail business. Like this is something which I think VNET has the history, but also has still thought the infrastructure to be able to service the customers more complicated, actually customer needs of the retail customers. It's not as simple as saying it's a retail customer, but using a colo model to service them. And so we're very confident on the fact that we have a distinct competitive advantage in that and we show in each of the quarters the ability to not only keep, but also then continues to add the retail customers to our base. But secondly, I think more importantly, this is what Samuel was referring to earlier on is the value-added services it's not just to sell retail customer colo service and the whole reason why it is a very attractive proposition for them to come VNET in the first place is the ability for them to come into a one-stop-shop a four sect of services. Not having to deal with 3, 4, 5 different vendors of the service. And I think that's a distinct advantage, which also then adds to the stickiness of the customer when people are coming into our data centers. The second part of question in terms of the customer demand pricing side, maybe Samuel, do you want to give a little bit of color to Albert.

Yes. First of all, I would say first and foremost, it is great to have a dual-core gross engine basic to help us. The IDC service providers to balance over, I would say resource and priorities. Resource are finite while the opportunities are infinite. And so when our peer companies in focuses switched to the retail to serve greedy indoors, our strategy. It is true that we're seeing a greater market opportunities in momentum in the retail front. IDC business overall, the wholesaler retails are shorter, like kind of -- different kind of a way to take that from a spectrum perspective. One focus on the quantity, while the other. Focus on the qualities. The wholesale business gives us a greater pretty predictability and also give us a scale. While the retail probably give us a very healthy margin, and also the stickiness of the customers. And if I agree, it is easier for retail business to go wholesale. While it is to provide for the wholesale to enter into the retail because you have to have the relationship with the customers, you have to have the track records, you have to have the great networking capabilities in companies. You have to have the whole bunch of value added services, and your sales structure compensation package skill set is going to be so different. And so, it's not just about a one-company want to go to retail. Actually, every single IDC service providers want to have the dual-core gross agent from my opinion. But it is hard. But again, I would say, in a given period of time, everybody want to do a wholesale business. But on the other hand, when there's historic coming to the wholesale front, and then the lot of company want to go to a retail as well. So I would say we're very fortunate to have the balanced strategy. And of course, as I mentioned earlier, we're super bullish about the future, but we're cautious about every single step we're taking, because resources to finding, so we want to make sure that we can optimize for the better visual return. So hopefully, that gives helper some of the additional color for the questions.

Thank you. Because of time, we'll now pick the last questions from Guohan Wang of Daiwa. Please go ahead.

Hello. Thanks management for the opportunity. My first question is regarding the deliver rates schedule. So we are being more cash sure about the near-term outlook. So as our capacity addition and more back in the loaded in 2022, and what's our view ongoing utilization ramp up, but you got kind of ramp up for 2022 and my second question about the recent moving progress and we see certain cases in Shanghai rate regions and also the other thinking originally and then we have -- what's our view on the impact of competitor deliver originated regions and we are expanding our hands to be a more the direct supplied customer mix or back for eight weeks to provide a more revenue mix on Shanghai and the Beijing region. Thanks.

Okay. Let me take the first question. Guohan thank you very much for the question itself. In terms of the capacity, yes I am expecting that more of the range of cabinets we're aiming to deliver will be in the latter part of the year. Next quarter we'll be able to give you, or give the market as in previous quarters, a clear guidance on how much of it we're expecting within first-half, and versus how much of it will be in the second half of the year. But it is very much in line with what was the case last year in that things will be more back-ended as we're taking a more cautious approach in terms of how quickly we deliver new capacity while we actually have our main focus is making sure that we ramp up the existing capacity. That leads to the second question, which is kind of what we're expecting in terms of the ramp in your utilization rate trend. I would say that you should see a dip, as I mentioned before, in the early part of year, and that will slowly then increase during the course of the year to the end part of the year as the ramp up continues across all of our different data centers. Samuel, do you want to take, I guess,the question -- I think there were some questions around sort of customers moving rates, ramp up progress?

No, I think you covered that pretty well, so no additional coming from me.

Ladies and gentlemen, that concludes our conference call today. Thank you for participating. You may now disconnect your lines.